simulating secondary organic aerosol from missing diesel ... · concluded that the contribution of...

Atmos. Chem. Phys., 16, 6453–6473, 2016www.atmos-chem-phys.net/16/6453/2016/doi:10.5194/acp-16-6453-2016© Author(s) 2016. CC Attribution 3.0 License.

Simulating secondary organic aerosol from missing diesel-relatedintermediate-volatility organic compound emissions during theClean Air for London (ClearfLo) campaignRiinu Ots1,2, Dominique E. Young3,a, Massimo Vieno2, Lu Xu4, Rachel E. Dunmore5, James D. Allan3,6, Hugh Coe3,Leah R. Williams7, Scott C. Herndon7, Nga L. Ng4,8, Jacqueline F. Hamilton5, Robert Bergström9,10, Chiara DiMarco2, Eiko Nemitz2, Ian A. Mackenzie11, Jeroen J. P. Kuenen12, David C. Green13, Stefan Reis2,14, andMathew R. Heal11School of Chemistry, University of Edinburgh, Edinburgh, UK2Natural Environment Research Council, Centre for Ecology & Hydrology, Penicuik, UK3School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Manchester, UK4School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, GA, USA5Wolfson Atmospheric Chemistry Laboratories, Department of Chemistry, University of York, York, UK6National Centre for Atmospheric Science, University of Manchester, Manchester, UK7Aerodyne Research, Inc., Billerica, MA, USA8School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA, USA9Swedish Meteorological and Hydrological Institute, Norrköping, Sweden10Department of Chemistry and Molecular Biology, University of Gothenburg, Gothenburg, Sweden11School of GeoSciences, University of Edinburgh, Edinburgh, UK12TNO, Department of Climate, Air and Sustainability, Utrecht, the Netherlands13MRC PHE Centre for Environment and Health, King’s College London, London, UK14University of Exeter Medical School, Knowledge Spa, Truro, UKanow at: Department of Environmental Toxicology, University of California, Davis, CA, USA

Correspondence to: Mathew R. Heal ([email protected]) and Riinu Ots ([email protected])

Received: 12 November 2015 – Published in Atmos. Chem. Phys. Discuss.: 18 January 2016Revised: 26 April 2016 – Accepted: 11 May 2016 – Published: 27 May 2016

Abstract. We present high-resolution (5 km× 5 km) atmo-spheric chemical transport model (ACTM) simulations ofthe impact of newly estimated traffic-related emissions onsecondary organic aerosol (SOA) formation over the UKfor 2012. Our simulations include additional diesel-relatedintermediate-volatility organic compound (IVOC) emissionsderived directly from comprehensive field measurements atan urban background site in London during the 2012 CleanAir for London (ClearfLo) campaign. Our IVOC emissionsare added proportionally to VOC emissions, as opposed toproportionally to primary organic aerosol (POA) as has beendone by previous ACTM studies seeking to simulate the ef-fects of these missing emissions. Modelled concentrationsare evaluated against hourly and daily measurements of or-

ganic aerosol (OA) components derived from aerosol massspectrometer (AMS) measurements also made during theClearfLo campaign at three sites in the London area. Ac-cording to the model simulations, diesel-related IVOCs canexplain on average ∼ 30 % of the annual SOA in and aroundLondon. Furthermore, the 90th percentile of modelled dailySOA concentrations for the whole year is 3.8 µg m−3, con-stituting a notable addition to total particulate matter. Moremeasurements of these precursors (currently not included inofficial emissions inventories) is recommended. During theperiod of concurrent measurements, SOA concentrations atthe Detling rural background location east of London weregreater than at the central London location. The model showsthat this was caused by an intense pollution plume with a

Published by Copernicus Publications on behalf of the European Geosciences Union.

6454 R. Ots et al.: Simulating SOA from missing diesel-related IVOC emissions in London

strong gradient of imported SOA passing over the rural loca-tion. This demonstrates the value of modelling for supportingthe interpretation of measurements taken at different sites orfor short durations.

1 Introduction

Ambient airborne particulate matter (PM) has diversesources and physicochemical properties. It affects the trans-port, transformation, and deposition of chemical species, andhas significant impacts on radiative forcing and on humanhealth (Pöschl, 2005; USEPA, 2009). The elemental and or-ganic carbon (EC and OC) components constitute a substan-tial proportion of total particle mass (USEPA, 2009; Putaudet al., 2010; AQEG, 2012). However, the characterization andsource apportionment of the organic component remains amajor challenge (Fuzzi et al., 2006; Hallquist et al., 2009;Jimenez et al., 2009). Understanding the sources of this or-ganic aerosol (OA) is important in order to devise effectivereduction strategies for ambient PM concentrations (Healet al., 2012).

Organic aerosol is typically a complex mixture of thou-sands of organic species, the majority of which are presentat low concentrations (less than a few ng m−3). Current lev-els of scientific understanding, instrumentation, and mod-elling capability mean that explicit measurement and mod-elling of all individual OA species is not feasible at present.Measurement of OA by on-line mass spectrometry, such aswith the Aerodyne aerosol mass spectrometer (AMS; Cana-garatna et al., 2007), and consideration of individual organicmarker ions coupled with multivariate statistical techniquessuch as positive matrix factorization (PMF; Paatero and Tap-per, 1994; Paatero, 1997) have facilitated the subdivisionof the OA component into empirical categories. These in-clude hydrocarbon-like organic aerosol (HOA), oxygenatedorganic aerosol (OOA, which can be further split into low-volatility and semi-volatile oxygenated organic aerosol: LV-OOA and SV-OOA), solid-fuel organic aerosol (SFOA) orbiomass burning organic aerosol (BBOA), cooking organicaerosol (COA), and a number of other categories (Ulbrichet al., 2009; Ng et al., 2010, 2011; Lanz et al., 2010; Younget al., 2015a). The SFOA factor includes biomass aerosolsthat are the so-called BBOA as well as OA from coal andcharcoal combustion (Allan et al., 2010).

Even allowing for the uncertainties in defining and mea-suring OA components, there is a general tendency for atmo-spheric chemical transport model (ACTM) simulations to un-derestimate observed amounts of OA and SOA. For example,the AeroCom (Aerosol Comparisons between Observationsand Models) project, which includes ∼ 30 global ACTMsand global circulation models (GCMs), has concluded thatthe amount of OA present in the atmosphere remains largelyunderestimated (Tsigaridis et al., 2014). Similarly, in an eval-

uation of seven global models, Pan et al. (2015) reported asystematic underestimation of OA over South Asia. Globalmodelling studies of SOA specifically have demonstratedhuge uncertainties (up to 10-fold) in total simulated SOAbudgets (Pye and Seinfeld, 2010; Spracklen et al., 2011;Jathar et al., 2011).

Several regional ACTM studies have also reported an un-derestimation of total OA (Simpson et al., 2007; Murphy andPandis, 2009; Hodzic et al., 2010; Aksoyoglu et al., 2011;Jathar et al., 2011; Bergström et al., 2012; Koo et al., 2014)and SOA (Hodzic et al., 2010; Shrivastava et al., 2011; Zhanget al., 2013; Fountoukis et al., 2014), with normalized meanbiases (NMBs) often in the range of −30 to −50 %. In somecases, this underestimation has been shown to be due to prob-lems with the underlying emission inventories, particularlyfor domestic wood burning in wintertime (Simpson et al.,2007; Denier van der Gon et al., 2015). There may also besources of biogenic secondary organic aerosol (BSOA) aris-ing from previously neglected VOC emissions such as thoseinduced by biotic stress (Berg et al., 2013; Bergström et al.,2014).

Currently, ACTMs cannot explicitly simulate all the ki-netic and thermodynamic processes associated with theevolving gas-phase chemistry of semi-volatile organic com-pounds and their partitioning to the particle phase (Don-ahue et al., 2014). A widely used heuristic parametriza-tion for simulating OA is the volatility basis set (Donahueet al., 2011, 2012). The volatility (in this case the satura-tion concentration at 298 K, C∗) of gas-phase organic com-pounds are sorted into bins: low-volatility organic com-pounds (LVOCs, C∗≤ 0.1 µg m−3; with no lower C∗, thiscategory also incorporates extremely low-volatility organiccompounds, ELVOCs), semi-volatile organic compounds(SVOCs, C∗= 1–103 µg m−3), intermediate-volatility or-ganic compounds (IVOCs, C∗= 104–106 µg m−3), andvolatile organic compounds (VOCs, C∗≥ 107). Thus, or-ganic compounds are distributed across a continuum fromparticles to gases. Under typical ambient conditions, allLVOCs, some of the SVOCs, and essentially none of theIVOCs or VOCs are in the condensed phase (Donahue et al.,2006).

Current emissions inventories, however, only report esti-mates for VOCs (C∗≥ 107 µg m−3) and for the particle frac-tion of the emissions of species with lower volatilities. Themain reason for this is that compounds with intermediatevolatility (SVOCs and IVOCs) are difficult to quantify, andthis is currently not routinely undertaken.

Robinson et al. (2007) and Shrivastava et al. (2008)estimated the mass of emitted IVOCs to be 1.5 timesthat of POA emissions. In their study, this addition ofIVOCs= 1.5×POA was applied to all sources of POA –from diesel to biomass burning. They based this estimationon chassis dynamometer tailpipe measurements by Schaueret al. (1999). Since then, several regional and global ACTMapplications have adopted this factor of 1.5 (e.g. Murphy

Atmos. Chem. Phys., 16, 6453–6473, 2016 www.atmos-chem-phys.net/16/6453/2016/

R. Ots et al.: Simulating SOA from missing diesel-related IVOC emissions in London 6455

and Pandis, 2009; Tsimpidi et al., 2010, 2016; Hodzic et al.,2010; Jathar et al., 2011; Fountoukis et al., 2011; Genberget al., 2011; Bergström et al., 2012; Zhang et al., 2013; Kooet al., 2014). A number of studies, including many of thosecited above reporting model underestimation of OA, havehighlighted the need for improved measurement and speci-ation of SVOCs and IVOCs and for these species to be re-ported in inventories.

Jathar et al. (2014) performed emissions and smog cham-ber experiments on SOA production from petrol and dieselvehicles. Diesel contains hydrocarbons with a higher car-bon number (C8–C20) than petrol (mainly C4–C10). The typ-ical method used for hydrocarbon analysis is gas chromatog-raphy (GC); however, as the carbon number increases, thenumber of potential structural isomers increases exponen-tially, meaning GC is unable to distinguish individual speciesin the intermediate-volatility range (Goldstein and Galbally,2007). The total carbon of this unresolved complex mixturewas estimated and Jathar et al. (2014) concluded that theseunspeciated organic gases dominate SOA production com-pared with SOA from the speciated precursors commonly in-cluded in emissions inventories (single-ring aromatics, iso-prene, terpenes, and large alkenes). Jathar et al. (2014) alsoperformed box-model simulations of the SOA budget for theUS, with the addition of unspeciated emissions based onmeasurements by Schauer et al. (1999), and concluded thatpetrol contributes much more to SOA than does diesel. Thisresult is similar to that of Bahreini et al. (2012), who, basedon measurements in the Los Angeles Basin, California (CA),concluded that the contribution of diesel emissions to SOAwas zero within measurement uncertainty. Conversely, Gen-tner et al. (2012) reported that diesel was responsible for65–90 % of vehicular-derived SOA based on measurementsof gas-phase organic carbon in the Caldecoff Tunnel, CA,and in Bakersfield, CA, and on estimations of SOA yields.The huge dissimilarity in these conclusions, even in the samestate in the US, emphasizes the need for continued researchinto petrol- and diesel-related SOA formation. Furthermore,the US and Europe have very different diesel vehicle pro-files: in the US, a negligible proportion of passenger carsare diesel (3 %), whilst on average across Europe 33 % ofpassenger cars are diesel and this proportion is increasing(Cames and Helmers, 2013). Globally, the demand for dieselfuel is increasing and by 2020 it is expected to overtake petrolas the principal transport fuel used worldwide (Exxon Mobil,2014).

In this study, we present new high-resolution simulationsof SOA formation in a 3-D ACTM model which includes ad-ditional diesel-related IVOC emissions derived directly fromcomprehensive field measurements of IVOCS and VOCs atan urban background site in central London (Dunmore et al.,2015) during the Clean Air for London (ClearfLo) cam-paign in 2012 (Bohnenstengel et al., 2014). Modelled con-centrations are compared with OA components derived byPMF analysis of AMS measurements during the same cam-

paign, including comparisons with the long-term measure-ments (full year) as well as the two month-long intensive ob-servation periods (IOPs) in winter and summer.

2 Methods

2.1 Model description

The EMEP4UK model is a regional application of the EMEPMSC-W (European Monitoring and Evaluation ProgrammeMeteorological Sythesizing Centre-West) model. The EMEPMSC-W model is a 3-D Eulerian model that has been usedfor both scientific studies and policy making in Europe. Adetailed description of the EMEP MSC-W model, includ-ing references to evaluation and application studies, is avail-able in Simpson et al. (2012), Schulz et al. (2013), and athttp://www.emep.int. The EMEP4UK model is described inVieno et al. (2010, 2014), and the model used here is basedon version v4.5.

The EMEP4UK model uses one-way nesting froma 50 km× 50 km greater European domain to a nested5 km× 5 km area covering the British Isles and parts of thenear continent. The model has 21 vertical levels, extend-ing from the ground to 100 hPa. The lowest vertical layeris ∼ 40 m thick, meaning that modelled surface concentra-tions represent the average for a 5 km× 5 km× 40 m gridcell. The model time step varies from 20 s (chemistry) to5 and 20 min for advection in the inner and outer domains,respectively. The chemical scheme used in this study isEMEP-EmChem09soa with the MARS equilibrium modulefor gas–aerosol partitioning of secondary inorganic aerosol(Binkowski and Shankar, 1995; Simpson et al., 2012).

The model was driven by output from the WeatherResearch and Forecasting (WRF) model (http://www.wrf-model.org, version 3.1.1) including data assimilation of6-hourly model meteorological reanalysis from the US Na-tional Center for Environmental Prediction (NCEP)/NationalCenter for Atmospheric Research (NCAR) Global ForecastSystem (GFS) at 1◦ resolution (NCEP, 2000). The WRFconfiguration was as follows: Lin Purdue for microphysics;Grell-3 for cumulus parametrization; Goddart Shortwave forradiation physics; and Yonsey University (YSU) for plane-tary boundary layer (PBL) height (see NCAR, 2008, for fur-ther information). This configuration is identical to that pre-sented in Vieno et al. (2010), where it is shown to performvery well in comparison with measurements. No further eval-uation is presented here.

2.2 Emissions

Gridded anthropogenic emissions of NOx (NO+NO2), SO2,NH3, CO, NMVOCs (non-methane VOCs), PM2.5 (PM withaerodynamic diameter < 2.5 µm) and PM10 (PM with aero-dynamic diameter < 10 µm) were obtained from NAEI (Na-tional Atmospheric Emissions Inventory, NAEI, 2013, for the

www.atmos-chem-phys.net/16/6453/2016/ Atmos. Chem. Phys., 16, 6453–6473, 2016

http://www.emep.int

http://www.wrf-model.org

http://www.wrf-model.org


Table 1. SNAP source sectors as specified in the emissions input tothe model (CEIP, 2015).

SNAP1 Combustion in energy and transformation industriesSNAP2 Residential and non-industrial combustionSNAP3 Combustion in manufacturing industrySNAP4 Production processesSNAP5 Extraction and distribution of fossil fuelsSNAP6 Solvent and other product useSNAP7 Road transportSNAP8 Other mobile sources and machinerySNAP9 Waste treatment and disposalSNAP10 Agriculture

UK and from CEIP (EMEP Centre on Emission Inventoriesand Projections; CEIP, 2015) for the rest of the model do-main. All emissions are apportioned across a standard set ofemission source sectors, following the sector structure de-fined in the Selected Nomenclature for Air Pollutant (SNAP;EEA, 2013; Table 1), consistently applied across the wholedomain.

Primary PM emissions reported as PM2.5 and PM10 inthe NAEI and CEIP were speciated into EC, OA from fos-sil fuel combustion, OA from domestic combustion, and re-maining primary PM by source sectors (using splits devel-oped by Kuenen et al., 2014, as in Fig. 1). Default specia-tion of NMVOC emissions into 14 surrogate groups was used(Simpson et al., 2012). International shipping emissions fromEntec UK Limited (now Amec Foster Wheeler) were used(Entec, 2010). The annual sectoral total emissions are tem-porally distributed to hourly resolution using hour-of-day,day-of-week, and monthly emission factors for each sourcesector as incorporated in the EMEP ACTM (Simpson et al.,2012). Daily emissions of all the aforementioned trace gasesand particles from natural fires were taken from the FireINventory from NCAR version 1.0 (FINNv1; Wiedinmyeret al., 2011). Monthly NOx emissions from in-flight aircraft,soil, and lightning, as well as biogenic emissions of dimethylsulfide (DMS), are included as described in Simpson et al.(2012). Biogenic emissions of isoprene and monoterpenesare calculated by the model for every grid cell and time step.Estimated emissions of wind-blown dust and sea salt are alsoincluded, but these have no impact on the model simulationsof OA (Simpson et al., 2012).

2.3 SOA production in the model

The EMEP MSC-W model uses the 1-D volatility basis set(VBS; Donahue et al., 2006) approach for SOA formation,ageing and phase partitioning. The implementation of theVBS framework within the model, including various optionsfor the treatment of volatility distributions and ageing reac-tions is described by Bergström et al. (2012).

In the model set-up used here, POA is treated as non-volatile and inert, as is currently assumed by emissions in-

SNAP10

SNAP9

SNAP8

SNAP7

SNAP6

SNAP5

SNAP4

SNAP3

SNAP2

SNAP1

0 10 20 30 40UK national total emissions, Gg/year

SFOA

HOA

EC

Other/Mineral PM

IVOCs

Figure 1. Annual UK PM2.5 emissions by SNAP sector (Table 1)as specified in the NAEI (for year 2012), with each sector split intoPOA (HOA or SFOA), EC, and remaining PM following Kuenenet al. (2014). The red bars are additional IVOCs (not included inofficial emission totals) that can be estimated as 1.5× the POA massin that sector. They are included in this plot to give an indication ofthe relative mass of IVOC additions that have been used in otherstudies.

ventories. Having POA be non-volatile allows us to betteridentify and isolate the SOA formed from our additionaldiesel IVOCs. Furthermore, it has been demonstrated byShrivastava et al. (2011) that a two-species VBS simulates anevolution of oxygen : carbon ratios (O : C) similar to the nine-species VBS approach. The two bins of Shrivastava et al.(2011) were of volatility 0.01 and 105, which, because mate-rial with the lower volatility is always completely in the par-ticle phase under ambient conditions, is similar to our non-volatile treatment of POA.

Five volatility bins (C∗= 0.1, 1, 10, 100, 1000 µg m−3)are used for SOA production and ageing. The SOA yieldsfor alkanes, alkenes, aromatics, isoprene, and terpenes un-der high- and low-NOx conditions were taken from Tsim-pidi et al. (2010). Note that Tsimpidi et al. (2010) reportedyields for the four VBS bins between 1 and 1000 µg m−3.In this work, the lowest VBS bin (0.1 µg m−3) is used forthe ageing reactions, as well as for SOA from the additionaldiesel IVOCs (explained in the next section). SOA from alka-nes, alkenes, and terpenes is assumed to have an initial or-ganic matter to organic carbon ratio (OM / OC ratio) of 1.7,while that from isoprene and aromatics is assumed to be 2.0and 2.1, respectively (Bergström et al., 2012; Chhabra et al.,2010). For comparison, HOA and SFOA were assumed tohave OM / OC ratios of 1.25 and 1.70, respectively (as inBergström et al., 2012, based on Aiken et al., 2008). Bothanthropogenic SOA (ASOA; from alkanes, alkenes, and aro-matics) and BSOA (from isoprene and terpenes) undergo at-mospheric ageing by the hydroxyl (OH) radical in the model(with rate coefficient of 4.0× 10−12 cm3 molecule−1 s−1;Lane et al., 2008), resulting in a shift into the next lowervolatility bin and a mass increase of 7.5 %.

A constant background OA of 0.4 µg m−3 is used to repre-sent the contribution of OA sources not explicitly included inthe model (e.g. oceanic sources or spores; Bergström et al.,



Table 2. Comparison of diesel and petrol NMVOCs in the UKNational Atmospheric Emissions Inventory (NAEI) with the urbanbackground ambient concentrations measured during the ClearfLowinter intensive observation period in London.

NAEI 2012 Measurements∗

(emission) (concentration)

Diesel (I)VOCs 8 Gg yr−1 107 µg m−3

Petrol VOCs 31 Gg yr−1 33 µg m−3

Diesel/petrol 0.26 3.2

∗ Dunmore et al. (2015).

2014). This background OA is assumed to be highly oxy-genated and is therefore included under modelled SOA whencomparing with observations (with an OM / OC ratio of 2.0it is also assumed to be non-volatile).

2.4 Additional IVOCs from diesel

Current emissions inventories report highly volatile an-thropogenic VOCs of C∗≥ 107 µg m−3 (Passant, 2002).However, diesel vehicles also produce substantial emis-sions of species with intermediate volatility in the range105≤C∗≤ 106 µg m−3 (IVOCs), as has been shown by Dun-

more et al. (2015) from measurements made at an ur-ban background site in central London during the ClearfLoproject.

In this study, aliphatic IVOC emissions from diesel vehi-cles were introduced into the model proportionally to on-roadtransport VOC emissions, using n-pentadecane (C15H32) asa surrogate for the following reasons. First, the amount ofalkenes in diesel fuel is low (< 5 %; Gentner et al., 2012),so an alkane is the most appropriate surrogate. Second, alln-alkanes up to n-dodecane were individually speciated andquantified during two month-long intensive observation pe-riods (IOPs) during the ClearfLo project and there werestrong correlations between all n-alkanes that have a pre-dominately diesel source (Dunmore et al., 2015). Third,the rate constant for the linear alkane is a reasonable rep-resentation of the rate constant for all the (unmeasured)branched and cyclic isomers, as demonstrated by Dunmoreet al. (2015) for the C12 n-alkane, dodecane. The bulk ofdiesel emissions, however, are likely to have higher car-bon numbers than were measured by a comprehensive two-dimensional gas chromatography (GC×GC) system (Dun-more et al., 2015). The rate coefficient for the reaction be-tween n-pentadecane and OH has been measured in a num-ber of studies (k= 2.07× 10−11 cm3 molecule−1 s−1; Atkin-son and Arey, 2003) unlike for the majority of branched iso-mers in this range. Furthermore, measurements of diesel fuelcomposition have shown that the average carbon number on apercentage weight basis was 14.94 (Gentner et al., 2012), so

SNAP 10

SNAP 9

SNAP 8

SNAP 7

SNAP 6

SNAP 5

SNAP 4

SNAP 3

SNAP 2

SNAP 1

0 100 200 300UK national total emissions, Gg year

Other/mix VOC emission

Petrol-VOCs emission

Diesel-VOCs emission

Added diesel IVOCs

–1

Figure 2. Annual UK NMVOC emissions by SNAP sector (Table 1)as specified in the NAEI (for the year 2012), with the SNAP7 emis-sions subdivided into petrol and diesel vehicles, and with the addi-tional diesel-associated IVOC emissions input to the model in thisstudy shown in red.

n-pentadecane was considered to be an appropriate surrogatefor diesel emissions in general.

In the NAEI, emissions from petrol vehicles dominate theNMVOC emissions from road traffic, but measurements dur-ing the ClearfLo winter IOP showed that NMVOCs assignedto diesel vehicles dominated traffic-related NMVOC concen-trations. The amount of pentadecane emitted in the modelwas therefore set to match the measured diesel-(I)VOC(IVOCs+VOCs) to petrol-VOC ratio (Table 2, Fig. 2). Thispentadecane addition was then applied to every country in themodel domain using the same factor as for the UK. This firstapproximation is justified because the fleet share of diesel ve-hicles in the UK is similar to the European average (∼ 30 %;EEA, 2010), but it can introduce errors for specific countries.

For the oxidation products of C15H32, SOA mass yieldswere taken from Presto et al. (2010): 0.044, 0.071, 0.41, and0.30 for the 0.1, 1, 10, and 100 µg m−3 bins, respectively(Presto et al., 2010, did not report a yield for the 1000 µg m−3

bin). These yields are reported for SOA with unit density(1 g cm−3). In this work, SOA density was assumed to be1.5 g cm−3 (Tsimpidi et al., 2010; Bergström et al., 2012) andthe yields were increased accordingly.

For the UK, our approach adds 90 Gg of diesel-IVOCemission for the year 2012 (Fig. 2). The 1.5×POA approach(Shrivastava et al., 2008, based on measurements by Schaueret al., 1999) would only add 31 Gg (Fig. 1). Part of thisdiscrepancy could be attributable to the different methodsand circumstances used to derive the additions (this work:5 weeks of ambient measurements in a megacity; previousestimate: tailpipe laboratory measurements using differentinstruments). Another possible reason for the differenceis an underestimate in POA emissions in the inventory;more POA would increase the amount of proportionallyadded IVOCs. However, Dunmore et al. (2015) show thatlower-carbon-number (and higher-volatility) NMVOCs mea-sured during the ClearfLo campaign were consistent withemissions estimates. This lends confidence to adding IVOCsproportionally to reported NMVOC emissions, rather than



proportionally to POA emissions. Nevertheless, we have alsoperformed a model run using the POA-based IVOC estimate,also including the semi-volatile treatment of POA. Theemitted semi-volatile POA (SVOCs) and 1.5×POA IVOCsare assigned to nine VBS bins – 0.03×POA, 0.06×POA,0.09×POA, 0.14×POA, 0.18×POA, 0.30×POA,0.40×POA, 0.50×POA, 0.80×POA to the bins 0.01–106,respectively – totalling 2.5×POA (Shrivastava et al., 2008).Both SVOCs and IVOCs then go through atmosphericageing with OH (k= 4.0× 10−11 cm3 molecule−1 s−1;Shrivastava et al., 2008). In this case, the additional IVOCswere calculated from POA from all sources, not just traffic-related. Note that in the UK, most of the additional IVOCsof the POA-based approach would come from SNAP2 (res-idential and non-industrial combustion emissions; Fig. 1):18 Gg, whereas only 5 Gg would be added to SNAP7 (roadtransport; and 8 Gg to remaining sectors). SVOCs andIVOCs that have undergone at least one ageing shift and arein the particulate phase are included under SOA (in additionto ASOA and BSOA from VOCs as in the Base case). Dueto the very different absolute amounts and source categories(the latter of which also leads to differences in the spatialpattern and temporal variation of the additional emissions),detailed comparison of the two different additions is notjustified, and only annual total OA component budgets ofthe different addition methodologies are presented.

2.5 Summary of model experiments

Three runs of the EMEP4UK modelling system were per-formed for 2012:

– Base: all anthropogenic emissions as in officially re-ported inventories; emissions of biogenic VOCs are cal-culated by the model for each advection time step.

– addDiesel: Base+ additional diesel IVOCs addedproportionally to NMVOC emissions from traf-fic (2.3×SNAP7). The additional IVOCs were treatedusing n-pentadecane as surrogate species. The semi-volatile VBS species formed after oxidation of n-pentadecane were treated in the same way as the ASOAspecies from VOC oxidation (the same ageing rate andmass increase due to oxygen addition; see Sect. 2.3).

– 1.5volPOA: semi-volatile treatment ofPOA+ additional IVOCs added proportionally toall POA emissions (1.5×POA; as in Shrivastava et al.,2008, based on measurements by Schauer et al., 1999).Inclusion of anthropogenic and biogenic VOCs as inBase.

2.6 Comparison with measurements

Modelled OA2.5 (OA with diameter < 2.5 µm) is comparedwith non-refractory submicron (NR-PM1) OA measured by

Figure 3. Locations of measurement sites used in this study. Lon-don North Kensington is an urban background site, and Harwelland Detling are rural background sites. Underlying map from©OpenStreetMap contributors.

Aerodyne AMS instruments at an urban background site incentral London and at two rural sites (Xu et al., 2016; Younget al., 2015a, b; Bohnenstengel et al., 2014; site locations areshown in Fig. 3). The error introduced to the comparison bythe different size fractions is believed to be small, as mea-surements at an urban background site in Birmingham, Eng-land have shown that 90 % of organic carbon in PM2.5 is inthe submicron fraction (Harrison and Yin, 2008).

Different types of AMS were deployed in this campaign.At the London North Kensington site a compact time-of-flight AMS (cToF-AMS) was deployed for a full calendaryear (January 2012–January 2013), and a high-resolutiontime-of-flight AMS (HR-ToF-AMS) was also deployed forthe IOPs at the same site. A HR-ToF-AMS was deployedin Detling during the winter IOP, and in Harwell during thesummer IOP. PMF analysis was applied to each of the datasets to apportion measured OA into different components(Ulbrich et al., 2009). A detailed description of the deriva-tion and optimization of the factors retrieved from the AMSdata at Detling can be found in Xu et al. (2016), at Lon-don North Kensington in Young et al. (2015a, b) (all threeof these analyses were performed with the PMF2 solver),and at Harwell in Di Marco et al. (2016) (using the ME-2solver). The OM / OC ratios for each of the PMF data setspresented in this study were calculated with the Improved-Ambient method from Canagaratna et al. (2015). A summaryof the instruments, measurement periods, and resolved PMFfactors is given in Table 3. As our emissions inventory doesnot include cooking OA (NAEI, 2013), this factor could notbe compared with the model.

When AMS measurements and their PMF apportionmentsare compared, some disagreement is observed, as shown forthe two instruments measuring at the same time at the samelocation at London North Kensington. This is in part due to



Table 3. AMS measurements and resolved PMF factors during the ClearfLo campaign and the allocation of the PMF factors to SOAfor comparison with model simulations. Site locations are shown in Fig. 3. Site names are abbreviated as follows: NK – London NorthKensington; DET – Detling; HAR – Harwell.

PMF factors

Period Site Dates (year 2012) Instrument Primary Secondary

Winter IOP NK 13 Jan–8 Feb HR-ToF-AMS HOA, SFOA1, SFOA2, COA OOADET 20 Jan–14 Feb HR-ToF-AMS HOA, SFOA OOA

Summer IOP NK 21 Jul–19 Aug HR-ToF-AMS HOA, COA, unknown SV-OOA, LV-OOAHAR 3 Aug–20 Aug HR-ToF-AMS HOA SV-OOA, LV-OOA, N-OOA

Annual NK 11 Jan–24 Jan (2013)∗ cToF-AMS HOA, SFOA∗∗, COA OOA1, OO2∗∗

∗ As the cToF-AMS was retuned before the summer IOP and retuned to the previous tuning at the end of the IOP, the subsequent data could not be used in the PMF analysis(see Young et al. (2015a) for details). However, for the purpose of the comparison in this study, data from the HR-ToF-AMS, deployed at the same site during the summerIOP, were used to fill in this period. ∗∗ PMF analysis revealed the SFOA and OOA2 factors were convolved due to their similar, strong diurnal cycles. Daily averages havebeen used to estimate their concentrations (Young et al., 2015a).

y = 0.342 + 0.807 × x, r = 0.95 y = 0.0567 + 1.08 × x, r = 0.92 y = 0.444 + 0.834 × x, r = 0.88 y = 0.79 + 0.736 × x, r = 0.77

(a) HOA (b) SFOA (c) COA (d) SOA

0

2

4

6

0

2

4

6

0

2

4

6

0

2

4

6

0 2 4 6 0 2 4 6 0 2 4 6 0 2 4 6Daily averaged HR-ToF-AMS, mg m3

Dai

ly a

vera

ged

cToF

-AM

S, m

g m

3

–

–

Figure 4. Scatter plots of PMF-derived OA component concentrations – (a) HOA, (b) SFOA, (c) COA, and (d) SOA – based on differentAMS instruments at the London North Kensington site during the winter IOP. The dashed lines are the 2 : 1, 1 : 1, and 1 : 2 lines.

the differences in the types of AMS used, where more chem-ical information is retrieved from the HR-ToF-AMS, whichcan subsequently lead to differences in the derived PMF fac-tors from the individual data sets. It should also be kept inmind that PMF was run on each of the full data sets, cov-ering a full year for the cToF-AMS and only four weeks foreach of the HR-ToF-AMS IOPs; thus, it is not necessarily ex-pected that the same PMF factors would be derived from thedifferent data sets. Nevertheless, strong correlations betweendaily averaged primary OA components from the two instru-ments deployed at the London North Kensington site duringthe winter IOP are observed (0.95, 0.92, and 0.88 for HOA,SFOA, and COA, respectively), with less strong correlationsfor SOA (0.77). Scatter plots of these PMF derived OA com-ponent concentrations resolved for the cToF-AMS data andHR-ToF-AMS are shown in Fig. 4. This inherent uncertaintyin the measurements constrains the expected correlation withthe model.

The following numerical metrics were used for modelevaluation:

– FAC2 (factor of 2): the proportion of modelled concen-trations that are within a factor of 2 of the measuredconcentrations.

– NMB: normalized mean bias.

– NMGE: normalized mean gross error, which is definedas

NMGE=

1n

n∑i=1|Mi −Oi |

O, (1)

where Mi is the ith modelled value, Oi is the cor-responding measured value, O is the mean measuredvalue, and n in the total number of observations.

– r: correlation coefficient.

– COE: coefficient of efficiency, which is defined as

COE= 1.0−

n∑i=1|Mi −Oi |

n∑i=1|Oi −O|

. (2)

A COE of 1 indicates perfect agreement between model andmeasurements. Although the COE does not have a lowerbound, a zero or negative COE implies that the model cannot



explain any of the variation in the observations (Legates andMcCabe, 2013).

Seasons are defined as follows: winter – December–January–February (DJF); spring – March–April–May (MAM); summer – June–July–August (JJA); andautumn – September–October–November (SON).

3 Results

The comparisons between the model results and measure-ments are presented in the following order. First, compar-isons are presented for primary OA, NOx , O3, and for sec-ondary inorganic aerosol (SIA) to give an overview of theoverall performance of the modelling system. Second, thehourly concentrations of SOA during the two IOPs are eval-uated, demonstrating the agreement between the model andmeasurements at high temporal resolution. Third, the year-long daily SOA concentrations are compared and the rela-tive impact of diesel VOCs on SOA production in London isshown. Fourth, modelled and measured OM / OC ratios areshown. Finally, annual total ASOA from our method and theprevious 1.5×POA approach are compared.

3.1 POA, NOx , O3, SIA: annual data set

Figure 5 shows the year-long comparison between the dailyaveraged model results and the cToF-AMS measurementsat the London North Kensington site. The model underesti-mates primary OA (HOA and SFOA) concentrations (NMBof−54 and−71 %, respectively) but shows good daily corre-lations (r values of 0.53 and 0.72, respectively). The underes-timation of HOA may be caused by a combination of lack ofmodel resolution (e.g. the minor road close to the measure-ment site can not be resolved with the 5 km grid), and under-estimation of PM emissions. Modelled NOx concentrationsare relatively less underestimated in comparison to measure-ments (NMB of−32 %, Fig. 6a), suggesting that HOA emis-sions may be more underestimated than the emissions ofNOx . Concentrations of secondary inorganic pollutants aresimulated well by the model in the gas phase (Fig. 6b, with aNMB of −1 % for ozone) and for inorganic PM constituents(Fig. 6c–d), with NMBs of 6 % for SO2−

4 , −12 % for NH+4 ,and −23 % for NO−3 .

3.2 Hourly comparison of secondary OA: summer IOP

Evaluation statistics between hourly measured and modelledSOA concentrations in July and August 2012 (summer IOP)show excellent agreement (Fig. 7). The values of r for theBase run were 0.67 and 0.55 at North Kensington and Har-well, respectively. The addDiesel experiment yields a mod-est improvement in the value of r at North Kensington(to 0.76) and a marked improvement in Harwell (to 0.74).The addDiesel run substantially improves the NMB for SOAat the Harwell and London North Kensington sites from

−32 to −5 % and from −35 to 0.1 %, respectively (Fig. 7).This means that ∼ 30 % of SOA at both sites during this pe-riod can be explained by the diesel IVOCs added into themodel using pentadecane as a surrogate. There is also markedimprovement of model–measurement COE values at the twosites (Harwell, 0.26 to 0.42; NK, 0.31 to 0.45). The improve-ment in NMGE is noticeable (Harwell, 54 to 43 %; NK, 59 to47 %), but smaller than the improvements in the other met-rics. It can be seen from the scatter plots in Fig. 8 that mostmodelled hourly SOA concentrations fall within a factor of2 of the measured concentrations (FAC2 for the addDieselexperiment is 78 % at Harwell and 62 % at NK).

Measured and modelled mean hour-of-day variations ofSOA concentrations are presented in Fig. 9, where it canbe seen that measured SOA concentrations do not have avery strong diurnal cycle. Interestingly, both sites exhibitdips in measured SOA concentrations in the morning andearly evening. Both measured and modelled SOA concen-trations in London North Kensington reach a maximum inthe afternoon, but SOA of the addDiesel experiment startsthis increase earlier than the measurements, meaning that ourASOA production from pentadecane might be too rapid.

During the summer IOP, there were two sustained episodesof increased SOA concentrations: 23 to 28 July and 9 to13 August (Fig. 7). Only London North Kensington had mea-surements during the first episode and the elevated concen-trations were well captured by the addDiesel simulation (in-cluding the highest peak of greater than 16 µg m−3: 27 July13:00 GMT, Fig. 10b). Daily averaged SOA maps (Fig. 11)suggest that this first episode arose from a combination ofSOA transported from Europe and SOA produced locallyin London. A region of elevated concentration around Lon-don exists within a general gradient of SOA from continen-tal Europe to southern England. Even daily averaged con-centrations are spatially variable during this episode mean-ing that inaccuracies in some of the modelled peaks can beattributed to uncertainties in the underlying meteorologicalmodel. Most of the modelled SOA during this episode wasof anthropogenic origin with the addDiesel run yielding asignificant portion of ASOA from pentadecane.

For the second sustained episode of high SOA concen-trations, from 9 to 13 August, several features remain sub-stantially underestimated even in the addDiesel run. For Har-well, the model captures two of the highest peaks (10 August22:00 GMT measured: 6.8 µg m−3, addDiesel: 8.5 µg m−3;12 August 12:00 GMT measured: 7.9 µg m−3, addDiesel:7.0 µg m−3), but for London North Kensington, the modelsimulates a minimum during the highest measured con-centration (10 August 05:00 GMT measured: 11.9 µg m−3,addDiesel: 2.0 µg m−3). The high concentrations during thefirst 2 days of this episode were very localized with horizon-tal widths of just tens of kilometres (Fig. 12a and b). Therewas a build-up of pollution caused by high pressure and lowboundary layer height (BLH), which led to production ofASOA in London. The high variability in the modelled con-



NMB = -54 %, NMGE = 56 %, r = 0.53, COE = 0.11(a) HOA

NMB = -71 %, NMGE = 72 %, r = 0.72, COE = 0.01(b) SFOA

0

1

2

3

4

5

6

0

1

2

3

4

5

6

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan

Dai

ly c

once

ntra

tion,

mg

m3

MeasuredBase

–

Figure 5. Time series of measured and modelled daily average concentrations of (a) HOA, and (b) SFOA at the London North Kensingtonurban background site, 2012, measured with the cToF-AMS (Table 3).

NMB = -32 %, NMGE = 36 %, r = 0.78, COE = 0.35(a) NOx

NMB = -1 %, NMGE = 24 %, r = 0.79, COE = 0.41(b) O3

NMB = 6 %, NMGE = 41 %, r = 0.73, COE = 0.32(c) SO42-

NMB = -12 %, NMGE = 50 %, r = 0.65, COE = 0.32(d) NH4+

NMB = -23 %, NMGE = 67 %, r = 0.57, COE = 0.34(e) NO3-

0

100

200

300

0

25

50

75

100

0.0

2.5

5.0

7.5

10.0

0

5

10

0

10

20

30

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan

Dai

ly c

once

ntra

tion,

mg

m3

MeasuredBase

–

Figure 6. Time series of measured and modelled daily average concentrations of (a) NOx (as NO2), (b) O3, (c) SO2−4 , (d) NH+4 , and

(d) NO−3 at the London North Kensington urban background site, 2012. Measurement data of NOx and O3 are from the UK Automated

Urban and Rural Network (AURN); SO2−4 , NH+4 , and NO−3 were measured with the cToF-AMS (Table 3).



NMB = -3 2 %, NMGE = 54 %, r = 0.55, COE = 0.26NMB = -5 %, NMGE = 43 %, r = 0.74, COE = 0.42

NMB = -35 %, NMGE = 59 %, r = 0.67, COE = 0.31NMB = 0.1 %, NMGE = 47 %, r = 0.76, COE = 0.45

(a) Harwell - Rural background

(b) London North Kensington - Urban background

2.5

5.0

7.5

0

5

10

15

Jul 2

1

Jul 2

2

Jul 2

3

Jul 2

4

Jul 2

5

Jul 2

6

Jul 2

7

Jul 2

8

Jul 2

9

Jul 3

0

Jul 3

1

Aug 0

1

Aug 0

2

Aug 0

3

Aug 0

4

Aug 0

5

Aug 0

6

Aug 0

7

Aug 0

8

Aug 0

9

Aug 1

0

Aug 1

1

Aug 1

2

Aug 1

3

Aug 1

4

Aug 1

5

Aug 1

6

Aug 1

7

Aug 1

8

Aug 1

9

Aug 2

0

Hou

rly S

OA

con

cent

ratio

n, m

g m

3

MeasuredBaseaddDiesel

–

Figure 7. Time series of measured and modelled hourly average concentrations at (a) the Harwell rural background site, and (b) the LondonNorth Kensington urban background site during the summer IOP. Note the different scales on the y axes.

Figure 8. Scatter plots of measured and modelled hourly SOAconcentrations during the summer 2012 IOP: (a) Base simulationat the Harwell rural background site; (b) Base simulation at theNorth Kensington urban background site; (c) addDiesel simulationat the Harwell rural background site; (d) addDiesel simulation atthe North Kensington urban background site. The straight lines arethe 2 : 1, 1 : 1, and 1 : 2 lines.

centrations (for example, the simulated minimum during themeasured maximum at North Kensington) is caused by theshifting of this narrow ASOA plume in space (Fig. 10b). On12 August, this episode was also subject to SOA contributionfrom Europe (Fig. 12d).

(a) Harwell (b) London North Kensington

0

2

4

6

8

0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21Hour of day

Mea

n S

OA

, mg

m3

Measured

addDiesel

–

Figure 9. Average hourly profiles of modelled (addDiesel experi-ment) and measured SOA during the summer IOP. Also shown arethe standard deviations for each mean value.

During the period of overlapping measurements at Har-well and North Kensington (3–18 August), both the mea-surements and the model agree with a modest rural to ur-ban increase. Average measured SOA concentrations were2.4 and 2.6 µg m−3 for Harwell and North Kensington,respectively, whilst average modelled concentrations were2.3 and 2.5 µg m−3 (for the addDiesel experiment).

3.3 Hourly comparison of secondary OA: winter IOP

Both the Detling and London North Kensington sites ex-hibit good model–measurement hourly correlation (r = 0.63and 0.64, addDiesel run; Fig. 13). The addDiesel run de-creases the NMB for SOA at these sites from −59 to−30 % for Detling, and from −24 to 8 % for London NorthKensington. This means that ∼ 30 % of SOA at these sitesduring this period can be explained by diesel IVOCs. InDetling, there is also a pronounced improvement in the COE,



Figure 10. Modelled (addDiesel experiment) hourly concentrationsof SOA at the time of the maximum measured hourly SOA value atthe London North Kensington site during the first and second SOAepisodes of the summer IOP. The white circles mark the measure-ment site locations – left: Harwell; right: London North Kensington.

from 0.10 to 0.31. In North Kensington, the COE was al-ready high but is increased from 0.27 to 0.30. It can be seenin Fig. 13 as well as Fig. 14 that lower concentrations of SOA(19—27 January) are overestimated by the model. This over-estimation is caused by the very simplified method of includ-ing missing sources of OA using a constant concentration of0.4 µg m−3 (which is assumed to be highly oxygenated and istherefore included under modelled SOA). As a constant, thisbackground OA does not currently go through atmosphericemission-removal processes in the model. However, the pe-riod in question exhibited snowfall, removing much of theaerosol (as can be seen from the very low concentrationsmeasured in both Detling and London North Kensington).Explicit inclusion of additional missing biogenic sources ofOA to the model is already part of ongoing development ofthe model and will be presented in future studies.

During the ClearfLo Winter IOP, measured SOA con-centrations were higher in Detling than in North Kensing-ton (Fig. 13). This is correctly captured by the simulationsand is caused by a steep positive gradient of concentrationsfrom southern England across to the near European continent(Fig. 15). The measured Detling / North Kensington SOA ra-tio (ratio of average concentrations for this period) was 1.8,while the modelled ratio was 1.1, so the model correctlysimulates the direction of the spatial gradient, but underes-timates its magnitude. For North Kensington, the model alsocaptures that SOA concentrations are lower on 5 Februarythan on 4 February. In Detling, however, measured concen-trations were higher on 5 February, which the model does notreproduce. During the night of 4–5 February, the wind wasvery strong (> 10 m s−1) and there was a small shift betweenthe measured wind direction and the wind direction input toEMEP4UK from WRF. As a consequence, the simulated pol-lution plume was shifted too much to the east (Fig. 15b),

causing the model–measurement discrepancy on this partic-ular occasion.

Even though the additional diesel IVOCs noticeably in-creased the modelled SOA concentrations during the win-ter IOP, there is still a marked underestimation of elevatedmeasured SOA concentrations during 15–19 January and30 January–4 February. During these periods, the observedtemperature was colder than the average temperature of thewinter IOP (Crilley et al., 2015) and peaks in measured SOAalso coincide with elevated concentrations of SFOA (Figs. 5band 13b). As our modelled SFOA is underestimated by a fac-tor of 4 (NMB of −72 %), it is likely that (i) SOA precur-sor VOC emissions from domestic heating are also under-estimated, and (ii) adding missing IVOCs from this emis-sion sector would contribute to the modelled SOA duringthese periods. It has been recently shown by Denier van derGon et al. (2015) that the emission factors used by differ-ent European countries for wood combustion PM emissions,even for the same appliance type, can differ by a factor of 5.They constructed a revised inventory, in which each coun-try’s emission was updated using an unified emission factor.This resulted in increases of PM (and estimated accompany-ing IVOC) emission estimates for most countries. Further-more, London is a smoke control area and therefore no resi-dential emissions of SFOA are assumed by the national emis-sions inventory. Recent studies have, however, suggested thatthere are indeed local sources of SFOA in London (Crilleyet al., 2015; Young et al., 2015a).

3.4 Daily and seasonal secondary OA: annual data set

Time series of daily averaged modelled and measured SOAconcentrations for the whole year are shown in Fig. 16. Ta-ble 4 gives daily modelled vs. measured SOA evaluationstatistics during different seasons at the North Kensingtonsite. Values for autumn are presented with and without thetwo extreme points (size of the data set n= 91 and n= 89).

For the daily model–measurement comparison, spring hasthe highest correlation (r = 0.85, both Base and addDiesel;Table 4). This can also be seen from the time series (Fig. 16:March–May) where both model simulations follow most ofthe measured peaks. The Base run r value for spring wasalready high, but, nevertheless, the addDiesel run shows amarked improvement for all other model evaluation statis-tics. FAC2 is increased by 10 %, COE is increased to 0.39,NMB is reduced by 35 %, and NMGE is reduced by 7 %.The NMGE of 38 % remaining in the addDiesel model run isprobably governed by uncertainties in meteorology, as wellas by uncertainties in the temporal and spatial variability ofemissions. During summer, the model captures the majorityof the periods of increased SOA mass well (e.g. 28 June, 22–29 July, 15 and 20 August; Fig. 16: June–August), but thereis some model underestimation when SOA concentrationswere lower (< 2 µg m−3). As for spring, the addDiesel exper-iment improves all model evaluation statistics. More detailed



Figure 11. Modelled (addDiesel experiment) daily average concentrations of SOA during the first SOA episode of the summer 2012 IOP.The white circle indicates the location of London North Kensington.

Figure 12. Modelled (addDiesel experiment) daily average concentrations of SOA during the second SOA episode of the summer 2012 IOP.The white circles mark the measurement site locations – left: Harwell; right: London North Kensington.

Figure 13. Time series of measured and modelled hourly average concentrations at (a) the Detling rural background site, and (b) the LondonNorth Kensington urban background site during the winter 2012 IOP. Note the different scales on the y axes.

hourly analysis of the SOA concentrations during the sum-mer IOP (end of July to August) was presented in Sect. 3.2.

The model performance is less good in autumn than dur-ing the other seasons. There are some days where the Basecase scenario overestimates measured SOA (23–25 Octo-ber, 21 and 24 November) with the addDiesel run increas-ing this further. During these days, particle nitrate (NO−3 )and ammonium (NH+4 ) are also substantially overestimated

by the model (Fig. 6). This suggests that the overestimationsare likely caused by errors occurring during this period inthe meteorological forecasts, e.g. missed rain events, ratherthan by uncertainties in the formation of secondary organicaerosol specifically.

The model evaluation statistics for autumn are strongly in-fluenced by the two modelled values on 23 and 24 October(Table 4). Removing these two values reduces the seasonal



Figure 14. Scatter plots of measured and modelled hourly SOAconcentrations during the winter 2012 IOP: (a) Base simulation atthe Detling rural background site; (b) Base simulation at the NorthKensington urban background site; (c) addDiesel simulation at theDetling rural background site; (d) addDiesel simulation at the NorthKensington urban background site. The straight lines are the 2 : 1,1 : 1, and 1 : 2 lines.

average SOA concentration modelled with the addDiesel runby 33 % (2.0 and 1.5 µg m−3 with and without these twopoints, respectively). Their combined influence on the annualaverage modelled concentration is 8 %, which is substantiallymore than any other points of the annual data set.

For the winter months, modelled concentrations in Januaryare much lower than measurements, whereas in February thetiming of several peaks is well reproduced and even overesti-mated by the addDiesel experiment. Detailed hourly analysisof the SOA concentrations during the winter IOP has beenpresented in Sect. 3.3. In December, measured SOA concen-trations were much lower than in January, and even thoughthe model captures the highest peak, there is some overesti-mation in the lowest range (< 0.5 µg m−3).

Figure 17 shows annually and seasonally averaged mea-sured and modelled SOA. The difference between the Baseand addDiesel experiments illustrate the impact of miss-ing IVOC emissions from diesel traffic on SOA formation.As was discussed before, and can be seen from Table 4,IVOC precursors from diesel vehicles reduce the NMB by∼ 30 %, which as an annual average is 0.6 µg m−3 of ad-ditional SOA. Moreover, the 90th percentile of daily av-eraged SOA concentrations of the addDiesel experiment is3.8 µg m−3 (which is similar to the measured 90th percentileof 3.2 µg m−3), whereas the 90th percentile of the Base casesimulation is 2.2 µg m−3. This means that (i), on 36 days

Table 4. Model–measurement comparison statistics for daily SOAat London North Kensington. Autumn is presented with and withoutthe two outliers (23 and 24 October n= 91 and 89, respectively).

Base addDiesel Base addDiesel

spring (MAM) summer (JJA)n (days) 91 86

FAC2 64 % 74 % 60 % 79 %NMB −35 % 0.1 % −34 % −5 %NMGE 45 % 38 % 48 % 39 %r 0.85 0.85 0.71 0.82COE 0.29 0.39 0.26 0.41

autumn (SON) winter (JFD)n (days) 89 81

FAC2 82 % 74 % 70 % 69 %NMB −2 % 58 % −28 % 6 %NMGE 52 % 96 % 47 % 61 %r 0.38 0.28 0.40 0.40COE −0.13 −1.07 0.21 −0.02

autumn (SON)n (days) 91

FAC2 80 % 73 %NMB 13 % 102 %NMGE 63 % 137 %r 0.58 0.54COE −0.30 −1.84

of the year, SOA is a notable component of PM (the an-nual average PM2.5 concentration limit value of the EuropeanUnion Directive 2008/50/EC is 25 µg m−3) and that (ii), dur-ing those days, the relative contribution to SOA from dieselIVOCs could be greater than 40 % (calculated as the differ-ence between SOA modelled with addDiesel and Base, rela-tive to addDiesel: (addDiesel-Base)/addDiesel). We note thatFig. 17a shows in the addDiesel simulation that the modelledBSOA+Background OA still makes up 53 % of the SOA,as an annual average. This value is based on the assignmentof the constant background OA in the model to natural SOA,which is what it is intended to represent. This may have someanthropogenic origin, and more research on the missing (orboundary condition) sources that this background constantrepresents is needed for accurate attribution of the biogenicvs. anthropogenic relative contributions.

3.5 OM / OC ratios

Measured OM / OC ratios for SOA were generally higherthan those modelled (1.99–2.34 vs. 1.88–1.97, Table 5).Nevertheless, the measured OM / OC ratio at London NorthKensington during the summer IOP was the lowest of themeasured range, 1.99, which is a close match to modelledSOA OM / OC ratio for that period, 1.97. Model perfor-mance for spring and summer was shown to be very good, but



Figure 15. Modelled (addDiesel experiment) daily average concentrations of SOA during the second SOA episode of the winter 2012 IOP.The white circles mark the measurement site locations – left: London North Kensington; right: Detling.

Figure 16. Time series of measured and modelled daily average SOA concentrations at the London North Kensington urban background site.The two outliers (23 and 24 October, included in the plot as labels) are excluded from the model evaluation statistics presented in the plot.

Table 5. Measured and modelled (addDiesel experiment) OM / OCratios. Site name abbreviations are given in Table 3.

Pollutant Site Period Meas. Mod.OM / OC OM / OC

HOA

NK winter IOP 1.25

1.25NK summer IOP 1.19NK annual 1.32HAR summer IOP 1.31DET winter IOP 1.45

SFOANK winter IOP 1.62

1.70NK annual 1.78DET winter IOP 1.64

SOA

NK winter IOP 2.03 1.88NK summer IOP 1.99 1.97NK annual 2.25 1.94HAR summer IOP 2.39 1.99DET winter IOP 2.34 1.86

it is possible that the missing SOA precursors in the coldermonths (from domestic heating) could yield SOA with higherinitial OM / OC ratios, thereby increasing the annual averagevalue. Furthermore, wintertime simulations of SOA in Parisby Fountoukis et al. (2016) also showed large underestima-tions, and they speculated that this could be pointing towards

an SOA formation process during periods of low photochem-ical activity that is currently not simulated in atmosphericchemical transport models.

3.6 Comparison to the previous (IVOCs= 1.5×POA)approach

Figure 18 shows the annual average HOA, SFOA, BSOA andbackground OA (BGND OA), and ASOA concentrations atLondon North Kensington modelled with different assump-tions for additional IVOC emissions. As was explained inSect. 2.4, for the UK, the addDiesel experiment adds 90 Ggof diesel-related IVOCs proportionally to road transportemissions (SNAP7), whereas the IVOCs= 1.5×POA ap-proach only adds 5 Gg to SNAP7 and another 26 Gg to othersectors (mainly to SNAP2: residential and non-industrialcombustion). Therefore, our approach creates a considerablylarger amount of SOA from IVOCs (and only from diesel-related IVOCs) than the previous method. The 1.5volPOAexperiment was undertaken using the semi-volatile treatmentof POA emissions. This means that the modelled ASOAfrom this experiment also includes aged semi-volatile POA,possibly giving it potential to create more ASOA than theBase or addDiesel experiments (the organic material addedto the model in the 1.5volPOA experiment is 1.0×POA (asSVOCs)+ 1.5×POA (IVOCs)= 2.5×POA as introducedby Robinson et al. (2007) and Shrivastava et al. (2008)). It



(a) Year (b) Spring (c) Summer (d) Autumn (e) Winter

0.0

0.5

1.0

1.5

2.0

2.5

Obs. Base addDiesel Obs. Base addDiesel Obs. Base addDiesel Obs. Base addDiesel Obs. Base addDiesel

Con

cent

ratio

n, m

g m

3

ASOA

BSOA

Background OA

SOA

–

Figure 17. Annually and seasonally averaged measured and modelled concentrations of SOA at the London North Kensington site.

can be seen from Fig. 18a and b that treating POA as semi-volatile leads to much lower concentrations than the non-volatile treatment (which already underestimates measuredconcentrations of HOA and SFOA by −54 and −71 %, re-spectively; Fig. 5). This is not surprising given that the semi-volatile treatment of POA assigns only 3 %+ 6 %+ 9 % ofthe POA to the three lowest volatility bins with saturationconcentrations of 0.01, 0.1, and 1 µg m−3, respectively (asgiven in Sect. 2.4). In a study in Mexico City, Shrivastavaet al. (2011) revised this treatment by assuming much highertotal semi-volatile and intermediate-volatility POA emis-sions: 7.5× the inventory emissions of (particulate) POA.This was justified by the fact that their emission factors ofPOA were derived from measurements at urban backgroundsites, but, following Robinson et al. (2007), two-thirds ofPOA would have evaporated by then. Recently, Shrivastavaet al. (2015) also used this factor of 7.5 in global simulations.Emission factors used in European inventories are, however,taken from tailpipe measurements with concentrations suffi-ciently high that most of the semi-volatiles should still be re-ported in the particulate phase. Therefore the further underes-timation of HOA and SFOA concentrations with the volatiletreatment could be due to a number of issues: (i) a system-atic underestimation of emissions, but for a different reasonthan in Shrivastava et al. (2011); (ii) the volatility of POA isoverestimated by Robinson et al. (2007), or (iii) the evapora-tion of semi-volatile POA emission is too rapid in the model(instantaneous in our set-up).

Figure 18c shows that the lower HOA and SFOA con-centrations lead to a very small negative change for the ab-sorptive partitioning of BSOA. Finally, it can be seen fromthe annual average concentrations of ASOA in Fig. 18d thatincluding aged SVOCs and IVOCs in the simulation dou-bles the modelled ASOA concentration compared to the Basecase scenario (ASOA from officially reported anthropogenicVOCs) but that the ASOA in our 1.5volPOA experiment isstill much lower than simulated with our addDiesel experi-ment.

(a) HOA (b) SFOA

(c) BSOA + BGND OA (d) ASOA0.00

0.25

0.50

0.75

1.00

0.00

0.25

0.50

0.75

1.00

Ann

ual a

vera

ge c

once

ntra

tion,

mg

m3 addDiesel

1.5volPOABase

–

Figure 18. Simulated annual and seasonal average concentrationsof OA components (BGND OA stands for background OA) for theLondon North Kensington site of three different model experiments:Base – all emissions as in officially reported emissions inventories,and POA is treated as non-volatile; 1.5volPOA – semi-volatile treat-ment of POA+ IVOC emissions added as 1.5×POA; addDiesel –Base+ IVOC emissions from diesel traffic added proportionally toVOC emissions from the on-road traffic source sector (SNAP7). Thelast two additions are as described in the main text.

4 Discussion

We show that ∼ 30 % of SOA in London could be producedfrom completely new estimates of diesel-related IVOC emis-sions that are not currently included in the emissions inven-tories. This is one of a very few studies where IVOC emis-sions are added proportionally to NMVOC emissions (as op-posed to addition proportionally to POA emissions). More-over, previous studies have added IVOCs proportionally toPOA from all sources, whereas this study focuses specificallyon the impact of diesel IVOCs from on-road traffic emissions(IVOCs= 2.3×SNAP7 VOCs). There is reason to believethat higher-volatility VOCs are better represented in currentemissions inventories than the emissions of PM. Also, theofficial inventories do not provide the individual contribu-



tion of POA to total PM. Therefore, the addition of IVOCsproportionally to NMVOCs may be better constrained thanthe POA-based approach used in studies so far. The addi-tional emissions are also tied directly to the relevant emissionsource category.

There are several possible uncertainties in our estimateof additional IVOCs, and subsequent SOA production andageing. As a first approximation, we added IVOCs to eachEuropean country based on our measurements in London.This was justified as the diesel usage in the UK is similarto the European average. Furthermore, different Europeancountries might be using different emissions factors for theirestimates of NMVOCs from petrol and diesel or have a dif-ferent average fleet age than the UK. It should be noted thattwo of the most populous countries in Europe – France andGermany – both have a higher diesel penetration than theUK ,and therefore for western central Europe our addition israther conservative.

It was seen from the hourly profiles at the London NorthKensington site during the summer IOP (Fig. 9b) that boththe model and the measurements exhibit a small diurnal cycle(peaking in the afternoon). Even though somewhat counter-intuitive (as most of the SOA chemistry is photochemicallydriven through reaction with the OH radical), an absence of astrong diurnal cycle of SOA has been seen in many Europeanstudies (Zhang et al., 2013; Fountoukis et al., 2014; Younget al., 2015a). A relatively small daytime increase in SOAcould be explained by the expansion of the boundary layerheight (Xu et al., 2015), as well as by contributions fromlong-range transport. PMF measurements of SOA in Mex-ico City, on the other hand, revealed a very strong diurnalcycle, peaking around midday (Shrivastava et al., 2011). Thefact that during the summer IOP our addDiesel experimentexhibits a slightly stronger diurnal cycle than the measure-ments (with daytime values slightly overestimated and night-time underestimated) indicates that the SOA yields could betoo high. We assumed an SOA density of 1.5 g cm−3 andincreased the yields linearly, as has been done in all otherACTM studies. Actually, increasing the assumed density ofSOA from the unit value (1 g cm−3) changes the total COA(condensed-phase OA) on the Odum mass yield plots (Odumet al., 1996) used to derive the yields from the chamber exper-iment. Therefore, increasing the yields linearly is not exactlycorrect (N. M. Donahue, personal communication, 2015) andfurther studies and refinement into the calculation of SOAyields and density would be beneficial.

We use an ageing rate of 4.0× 10−12 cm3 molecule−1 s−1

for both ASOA and BSOA (Lane et al., 2008). This is slowerthan has been used in some other studies (for example,Tsimpidi et al., 2010, use 4.0× 10−11 cm3 molecule−1 s−1,which is 10 times faster, and Fountoukis et al., 2011, use1.0× 10−11 cm3 molecule−1 s−1, which is 2.5 times faster).A combination of lower initial SOA yields but slightly higherageing rates could possibly flatten the diurnal cycle of ourmodelled SOA, matching the measurements better. There-

fore, an improvement for the detailed, hourly evolution couldbe achieved by a sensitivity study of these yields and ageingrates. This does not, however, change the main scope and re-sults of this paper which illustrate the relative impact of thediesel IVOCs on SOA formation.

In the current set-up of the EMEP model, only two PMsize fractions are simulated – PM2.5 and PM2.5−10 – becauseonly two fractions are included in the emissions inventoriesfor PM used in this study. Even though on an annual basis90 % of OC2.5 is in the submicron (OC1) range (Sect. 2.6),the comparison between a modelled OC2.5 and a measuredOC1 could be introducing larger errors during specific daysor hours. Therefore, as AMS measurements become moreprevalent, emissions inventories should be reported for allthree size classes, PM1, PM1−2.5, and PM2.5−10. This wouldallow the model to partition SOA into the corresponding frac-tions, making the direct comparison of modelled SOA1 tomeasured SOA1 possible.

We showed that treating POA as semi-volatile and let-ting it evaporate led to a great underestimation of HOA andSFOA concentrations compared to measurements at the Lon-don North Kensington urban background site. As has beenhighlighted by a number studies before us (listed in the In-troduction), we would also emphasize that a major source ofuncertainty in OA modelling is the volatility of primary emis-sions, an issue that is currently not addressed by official emis-sions inventories. In our experiment of semi-volatile POA(denoted 1.5volPOA), IVOCs were included from all sourcesectors. This experiment simulated substantially less ASOAthan our addition of IVOCs associated with just the trafficsource sector. This means that a combination of the POA-based IVOCs and our addition of diesel IVOCs proportionalto NMVOCs would not create a substantial overestimation ofSOA concentrations compared to measurements. Neverthe-less, further modelling studies (including different assump-tions regarding ageing rates, fragmentation, and yields) aswell as more measurements of IVOC emissions from differ-ent sources are clearly necessary.

In the evaluation of modelled and measured SOA, it wasshown that some of the uncertainties in the modelled concen-trations are caused by errors in modelled wind vectors. Nev-ertheless, the underlying meteorological model works well(as demonstrated by comparisons of different pollutants forthe whole calendar year), and overall the errors caused bymeteorology are believed to be relatively smaller than thoseintroduced by emissions (amount, volatility, composition), orSOA yields and ageing rates.

5 Conclusions

This study presents annual time series of new high-resolutionsimulations of SOA formation over the UK (using theEMEP4UK Eulerian atmospheric chemical transport model)that include diesel-related IVOC emissions not currently in-



cluded in the emissions inventory. The derivation of themagnitude of these additional emissions of SOA precursorsand the evaluation of the model simulated SOA were bothbased on measurements made during the Clean Air for Lon-don (ClearfLo) campaign in 2012. The IVOC emissions wereadded in proportion to the VOC emissions from the specif-ically relevant on-road traffic source, in contrast to previousstudies that have added IVOCs proportionally to primary or-ganic aerosol (POA) emissions from all POA sources. Mod-elled concentrations of SOA were compared with positivematrix factorization (PMF) analyses of aerosol mass spec-trometer (AMS) measurements at a central London urbanbackground location (North Kensington) and at the Detlingand Harwell rural background locations outside of London.

The model performance in comparison to relatively morewell-known components of air pollution, such as NOx , O3,and secondary inorganic aerosol, was shown to be very good,providing confidence in the prediction skill of the ACTMsystem used. Modelled concentrations of SOA were evalu-ated in four groups: (i) hourly comparison during a summerIOP (intensive observation period), (ii) hourly comparisonduring a winter IOP, (iii) daily comparison for a full cal-endar year (including seasonal statistics), and (iv) compar-ison of OM / OC ratios of all apportioned OA components.To our knowledge, this is the first study where modelled OAcomponents are compared with a year-long data set of PMF-apportioned AMS measurements.

During the period of concurrent measurements at all loca-tions, SOA concentrations at the Detling rural background lo-cation were greater than at the central London location. Themodel showed that this was caused by an intense pollutionplume with a strong gradient of SOA from mainland Europepassing over the rural location and demonstrates how shortperiods of measurements can give a different picture com-pared with longer-term measurements, as well as the valueof atmospheric chemistry transport modelling for supportingthe interpretation of measurements taken at different sites orfor short durations.

The model simulations show that these estimates of diesel-related IVOCs could explain on average∼ 30 % of the annualSOA in and around London. The 90th percentile of modelleddaily SOA concentrations at the urban background site forthe whole year was 3.8 µg m−3, and the influence of miss-ing diesel-related IVOC precursors was even greater on highpercentile SOA days than its contribution to annual averageSOA. The magnitudes of these contributions to SOA providestrong additional support for the need to undertake furtherrefinement of the amount and speciation of these precursoremissions for inclusion in official emissions inventories.

Data availability

Processed measurement data used in this study are avail-able through the ClearfLo project archive at the British At-mospheric Data Centre (http://badc.nerc.ac.uk/browse/badc/

clearflo). The model data (input, code, relevant output) arearchived at the University of Edinburgh, and Centre for Ecol-ogy&Hydrology and are available on request.

Acknowledgements. The authors acknowledge the UK Departmentfor Environment, Food and Rural Affairs (Defra) and the De-volved Administrations for several projects: development of theEMEP4UK model (AQ0727); access to the AURN data, which wereobtained from uk-air.defra.gov.uk and are subject to Crown 2014copyright, Defra, licensed under the Open Government Licence(OGL); and partial support for the aerosol measurements. Partialsupport for the EMEP4UK modelling from the European Commis-sion FP7 ECLAIRE project is gratefully acknowledged. This workwas supported in part by the UK Natural Environment ResearchCouncil (NERC) ClearfLo project (grant ref. NE/H008136/1).Riinu Ots was supported by a PhD studentship (Universityof Edinburgh and NERC-CEH contract 587/NEC03805). Do-minique E. Young was supported by a NERC PhD studentship(ref. NE/I528142/1). Rachel E. Dunmore was supported by aNERC PhD studentship (ref. NE/J500197/1). Nga L. Ng, Lu Xu,Leah R. Williams, and Scott C. Herndon were supported by theUS Department of Energy (grant no. DE-SC000602). The authorswould like to thank David Simpson for helpful advice about theEMEP model.

NCAR command language (NCL) was used to produce themaps (NCAR, 2015), and R, openair, and ggplot2 were used forthe analysis and all other plots (R Core Team, 2014; Carslaw andRopkins, 2012; Wickham, 2009).

Edited by: M. Kanakidou

References

Aiken, A. C., DeCarlo, P. F., Kroll, J. H., Worsnop, D. R., Huffman,J. A., Docherty, K. S., Ulbrich, I. M., Mohr, C., Kimmel, J. R.,Sueper, D., Sun, Y., Zhang, Q., Trimborn, A., Northway, M., Zie-mann, P. J., Canagaratna, M. R., Onasch, T. B., Alfarra, M. R.,Prevot, A. S. H., Dommen, J., Duplissy, J., Metzger, A., Bal-tensperger, U., and Jimenez, J. L.: O / C and OM / OC Ratios ofPrimary, Secondary, and Ambient Organic Aerosols with High-Resolution Time-of-Flight Aerosol Mass Spectrometry, Environ.Sci. Technol., 42, 4478–4485, doi:10.1021/es703009q, 2008.

Aksoyoglu, S., Keller, J., Barmpadimos, I., Oderbolz, D., Lanz, V.A., Prévôt, A. S. H., and Baltensperger, U.: Aerosol modelling inEurope with a focus on Switzerland during summer and winterepisodes, Atmos. Chem. Phys., 11, 7355–7373, doi:10.5194/acp-11-7355-2011, 2011.

Allan, J. D., Williams, P. I., Morgan, W. T., Martin, C. L., Flynn, M.J., Lee, J., Nemitz, E., Phillips, G. J., Gallagher, M. W., and Coe,H.: Contributions from transport, solid fuel burning and cook-ing to primary organic aerosols in two UK cities, Atmos. Chem.Phys., 10, 647–668, doi:10.5194/acp-10-647-2010, 2010.

AQEG: Report: Fine Particulate Matter (PM2.5) in the United – De-fra, UK, http://uk-air.defra.gov.uk/library/reports?report_id=727(last access: 24 May 2016), 2012.


http://badc.nerc.ac.uk/browse/badc/clearflo

http://badc.nerc.ac.uk/browse/badc/clearflo

http://dx.doi.org/10.1021/es703009q

http://dx.doi.org/10.5194/acp-11-7355-2011



http://uk-air.defra.gov.uk/library/reports?report_id=727


Atkinson, R. and Arey, J.: Atmospheric Degradation ofVolatile Organic Compounds, Chem. Rev., 103, 4605–4638,doi:10.1021/cr0206420, 2003.

Bahreini, R., Middlebrook, A. M., de Gouw, J. A., Warneke, C.,Trainer, M., Brock, C. A., Stark, H., Brown, S. S., Dube, W. P.,Gilman, J. B., Hall, K., Holloway, J. S., Kuster, W. C., Perring,A. E., Prevot, A. S. H., Schwarz, J. P., Spackman, J. R., Szidat,S., Wagner, N. L., Weber, R. J., Zotter, P., and Parrish, D. D.:Gasoline emissions dominate over diesel in formation of sec-ondary organic aerosol mass, Geophys. Res. Lett., 39, L06805,doi:10.1029/2011GL050718, 2012.

Berg, A. R., Heald, C. L., Huff Hartz, K. E., Hallar, A. G., Med-dens, A. J. H., Hicke, J. A., Lamarque, J., and Tilmes, S.: Theimpact of bark beetle infestations on monoterpene emissions andsecondary organic aerosol formation in western North America,Atmos. Chem. Phys., 13, 3149–3161, doi:10.5194/acp-13-3149-2013, 2013.

Bergström, R., Denier van der Gon, H. A. C., Prévôt, A. S., Yttri,K. E., and Simpson, D.: Modelling of organic aerosols over Eu-rope (2002–2007) using a volatility basis set (VBS) framework:application of different assumptions regarding the formation ofsecondary organic aerosol, Atmos. Chem. Phys., 12, 8499–8527,doi:10.5194/acp-12-8499-2012, 2012.

Bergström, R., Hallquist, M., Simpson, D., Wildt, J., and Mentel, T.F.: Biotic stress: a significant contributor to organic aerosol in Eu-rope?, Atmos. Chem. Phys., 14, 13643–13660, doi:10.5194/acp-14-13643-2014, 2014.

Binkowski, F. S. and Shankar, U.: The Regional Particulate MatterModel: 1. Model description and preliminary results, J. Geophys.Res.-Atmos., 100, 26191–26209, doi:10.1029/95JD02093, 1995.

Bohnenstengel, S. I., Belcher, S. E., Aiken, A., Allan, J. D., Allen,G., Bacak, A., Bannan, T. J., Barlow, J. F., Beddows, D. C. S.,Bloss, W. J., Booth, A. M., Chemel, C., Coceal, O., Di Marco,C. F., Dubey, M. K., Faloon, K. H., Fleming, Z. L., Furger, M.,Gietl, J. K., Graves, R. R., Green, D. C., Grimmond, C. S. B.,Halios, C. H., Hamilton, J. F., Harrison, R. M., Heal, M. R.,Heard, D. E., Helfter, C., Herndon, S. C., Holmes, R. E., Hop-kins, J. R., Jones, A. M., Kelly, F. J., Kotthaus, S., Langford,B., Lee, J. D., Leigh, R. J., Lewis, A. C., Lidster, R. T., Lopez-Hilfiker, F. D., McQuaid, J. B., Mohr, C., Monks, P. S., Nemitz,E., Ng, N. L., Percival, C. J., Prévôt, A. S. H., Ricketts, H. M. A.,Sokhi, R., Stone, D., Thornton, J. A., Tremper, A. H., Valach,A. C., Visser, S., Whalley, L. K., Williams, L. R., Xu, L., Young,D. E., and Zotter, P.: Meteorology, Air Quality, and Health inLondon: The ClearfLo Project, B. Am. Meteorol. Soc., 96, 779–804, doi:10.1175/BAMS-D-12-00245.1, 2014.

Cames, M. and Helmers, E.: Critical evaluation of the Europeandiesel car boom – global comparison, environmental effectsand various national strategies, Environ. Sci. Europe, 25, 1–22,doi:10.1186/2190-4715-25-15, 2013.

Canagaratna, M. R., Jayne, J., Jimenez, J., Allan, J., Alfarra, M.,Zhang, Q., Onasch, T., Drewnick, F., Coe, H., Middlebrook, A.,Delia, A., Williams, L., Trimborn, A., Northway, M., DeCarlo, P.,Kolb, C., Davidovits, P., and Worsnop, D.: Chemical and micro-physical characterization of ambient aerosols with the aerodyneaerosol mass spectrometer, Mass Spectrom. Rev., 26, 185–222,doi:10.1002/mas.20115, 2007.

Canagaratna, M. R., Jimenez, J. L., Kroll, J. H., Chen, Q., Kessler,S. H., Massoli, P., Hildebrandt Ruiz, L., Fortner, E., Williams, L.

R., Wilson, K. R., Surratt, J. D., Donahue, N. M., Jayne, J. T.,and Worsnop, D. R.: Elemental ratio measurements of organiccompounds using aerosol mass spectrometry: characterization,improved calibration, and implications, Atmos. Chem. Phys., 15,253–272, doi:10.5194/acp-15-253-2015, 2015.

Carslaw, D. C. and Ropkins, K.: openair – An R package for airquality data analysis, Environ. Model. Softw., 27–28, 52–61,doi:10.1016/j.envsoft.2011.09.008, 2012.

CEIP: WebDab EMEP database: Emissions as used in EMEP mod-els, http://www.ceip.at/ms/ceip_home1/ceip_home/webdab_emepdatabase/, last access: 15 July 2015.

Chhabra, P. S., Flagan, R. C., and Seinfeld, J. H.: Elemental analysisof chamber organic aerosol using an aerodyne high-resolutionaerosol mass spectrometer, Atmos. Chem. Phys., 10, 4111–4131,doi:10.5194/acp-10-4111-2010, 2010.

Crilley, L. R., Bloss, W. J., Yin, J., Beddows, D. C. S., Harrison, R.M., Allan, J. D., Young, D. E., Flynn, M., Williams, P., Zotter, P.,Prevot, A. S. H., Heal, M. R., Barlow, J. F., Halios, C. H., Lee, J.D., Szidat, S., and Mohr, C.: Sources and contributions of woodsmoke during winter in London: assessing local and regional in-fluences, Atmos. Chem. Phys., 15, 3149–3171, doi:10.5194/acp-15-3149-2015, 2015.

Denier van der Gon, H. A. C., Bergström, R., Fountoukis, C., Jo-hansson, C., Pandis, S. N., Simpson, D., and Visschedijk, A.J. H.: Particulate emissions from residential wood combustionin Europe – revised estimates and an evaluation, Atmos. Chem.Phys., 15, 6503–6519, doi:10.5194/acp-15-6503-2015, 2015.

Di Marco, C. F., Nemitz, E., Twigg, M. M., Langford, B., Lingard,J. J. N., and Ritchie, S.: Chemical characterisation of the aerosolat a background site in southern England with emphasis on theorganic fraction and nitrogen compounds, in preparation, 2016.

Donahue, N. M., Robinson, A. L., Stanier, C. O., and Pandis,S. N.: Coupled Partitioning, Dilution, and Chemical Aging ofSemivolatile Organics, Environ. Sci. Technol., 40, 2635–2643,doi:10.1021/es052297c, 2006.

Donahue, N. M., Epstein, S. A., Pandis, S. N., and Robinson, A.L.: A two-dimensional volatility basis set: 1. organic-aerosolmixing thermodynamics, Atmos. Chem. Phys., 11, 3303–3318,doi:10.5194/acp-11-3303-2011, 2011.

Donahue, N. M., Kroll, J. H., Pandis, S. N., and Robinson, A. L.:A two-dimensional volatility basis set – Part 2: Diagnostics oforganic-aerosol evolution, Atmos. Chem. Phys., 12, 615–634,doi:10.5194/acp-12-615-2012, 2012.

Donahue, N. M., Robinson, A. L., Trump, E. R., Riipinen, I.,and Kroll, J. H.: Volatility and Aging of Atmospheric OrganicAerosol, in: Atmospheric and Aerosol Chemistry, no. 339 in Top-ics in Current Chemistry, edited by: McNeill, V. F. and Ariya,P. A., Springer, Berlin, Heidelberg, 97–143, 2014.

Dunmore, R. E., Hopkins, J. R., Lidster, R. T., Lee, J. D., Evans,M. J., Rickard, A. R., Lewis, A. C., and Hamilton, J. F.:Diesel-related hydrocarbons can dominate gas phase reactivecarbon in megacities, Atmos. Chem. Phys., 15, 9983–9996,doi:10.5194/acp-15-9983-2015, 2015.

EEA: Dieselisation in the European Economic Area,Tech. rep., http://www.eea.europa.eu/data-and-maps/figures/dieselisation-in-the-eea (last access: 24 May 2016), 2010.

EEA: EMEP/EEA air pollutant emission inventory guide-book – 2013, http://www.eea.europa.eu/publications/emep-eea-guidebook-2013 (last access: 24 May 2016), 2013.


http://dx.doi.org/10.1021/cr0206420

http://dx.doi.org/10.1029/2011GL050718






http://dx.doi.org/10.1029/95JD02093

http://dx.doi.org/10.1175/BAMS-D-12-00245.1

http://dx.doi.org/10.1186/2190-4715-25-15

http://dx.doi.org/10.1002/mas.20115


http://dx.doi.org/10.1016/j.envsoft.2011.09.008

http://www.ceip.at/ms/ceip_home1/ceip_home/webdab_emepdatabase/

http://www.ceip.at/ms/ceip_home1/ceip_home/webdab_emepdatabase/





http://dx.doi.org/10.1021/es052297c




http://www.eea.europa.eu/data-and-maps/figures/dieselisation-in-the-eea

http://www.eea.europa.eu/data-and-maps/figures/dieselisation-in-the-eea

http://www.eea.europa.eu/publications/emep-eea-guidebook-2013

http://www.eea.europa.eu/publications/emep-eea-guidebook-2013


Entec: UK ship emissions inventory, final report, Tech. rep., crowncopyright, http://uk-air.defra.gov.uk/assets/documents/reports/cat15/1012131459_21897_Final_Report_291110.pdf, (lastaccess: 24 May 2016), 2010.

Exxon Mobil: The Outlook for Energy: A view to 2040,Tech. rep., Exxon Mobil Corporation, Texas, available at: http://cdn.exxonmobil.com/~/media/global/files/outlook-for-energy/2015-outlook-for-energy_print-resolution.pdf (last access:24 May 2016), 2014.

Fountoukis, C., Racherla, P. N., Denier van der Gon, H. A. C.,Polymeneas, P., Charalampidis, P. E., Pilinis, C., Wiedensohler,A., Dall’Osto, M., O’Dowd, C., and Pandis, S. N.: Evaluationof a three-dimensional chemical transport model (PMCAMx)in the European domain during the EUCAARI May 2008 cam-paign, Atmos. Chem. Phys., 11, 10331–10347, doi:10.5194/acp-11-10331-2011, 2011.

Fountoukis, C., Megaritis, A. G., Skyllakou, K., Charalampidis,P. E., Pilinis, C., Denier van der Gon, H. A. C., Crippa, M.,Canonaco, F., Mohr, C., Prévôt, A. S. H., Allan, J. D., Poulain, L.,Petäjä, T., Tiitta, P., Carbone, S., Kiendler-Scharr, A., Nemitz, E.,O’Dowd, C., Swietlicki, E., and Pandis, S. N.: Organic aerosolconcentration and composition over Europe: insights from com-parison of regional model predictions with aerosol mass spec-trometer factor analysis, Atmos. Chem. Phys., 14, 9061–9076,doi:10.5194/acp-14-9061-2014, 2014.

Fountoukis, C., Megaritis, A. G., Skyllakou, K., Charalampidis,P. E., Denier van der Gon, H. A. C., Crippa, M., Prévôt, A.S. H., Fachinger, F., Wiedensohler, A., Pilinis, C., and Pan-dis, S. N.: Simulating the formation of carbonaceous aerosolin a European Megacity (Paris) during the MEGAPOLI sum-mer and winter campaigns, Atmos. Chem. Phys., 16, 3727–3741,doi:10.5194/acp-16-3727-2016, 2016.

Fuzzi, S., Andreae, M. O., Huebert, B. J., Kulmala, M., Bond,T. C., Boy, M., Doherty, S. J., Guenther, A., Kanakidou, M.,Kawamura, K., Kerminen, V., Lohmann, U., Russell, L. M., andPöschl, U.: Critical assessment of the current state of scien-tific knowledge, terminology, and research needs concerning therole of organic aerosols in the atmosphere, climate, and globalchange, Atmos. Chem. Phys., 6, 2017–2038, doi:10.5194/acp-6-2017-2006, 2006.

Genberg, J., Hyder, M., Stenström, K., Bergström, R., Simpson,D., Fors, E. O., Jönsson, J. R., and Swietlicki, E.: Sourceapportionment of carbonaceous aerosol in southern Sweden,Atmos. Chem. Phys., 11, 11387–11400, doi:10.5194/acp-11-11387-2011, 2011.

Gentner, D. R., Isaacman, G., Worton, D. R., Chan, A. W. H., Dall-mann, T. R., Davis, L., Liu, S., Day, D. A., Russell, L. M., Wil-son, K. R., Weber, R., Guha, A., Harley, R. A., and Goldstein,A. H.: Elucidating secondary organic aerosol from diesel andgasoline vehicles through detailed characterization of organiccarbon emissions, P. Natl. Acad. Sci. USA, 109, 18318–18323,doi:10.1073/pnas.1212272109, 2012.

Goldstein, A. H. and Galbally, I. E.: Known and Unexplored Or-ganic Constituents in the Earth’s Atmosphere, Enviro. Sci. Tech-nol., 41, 1514–1521, doi:10.1021/es072476p, 2007.

Hallquist, M., Wenger, J. C., Baltensperger, U., Rudich, Y., Simp-son, D., Claeys, M., Dommen, J., Donahue, N. M., George, C.,Goldstein, A. H., Hamilton, J. F., Herrmann, H., Hoffmann, T.,Iinuma, Y., Jang, M., Jenkin, M. E., Jimenez, J. L., Kiendler-

Scharr, A., Maenhaut, W., McFiggans, G., Mentel, T. F., Monod,A., Prévôt, A. S. ., Seinfeld, J. H., Surratt, J. D., Szmigielski, R.,and Wildt, J.: The formation, properties and impact of secondaryorganic aerosol: current and emerging issues, Atmos. Chem.Phys., 9, 5155–5236, doi:10.5194/acp-9-5155-2009, 2009.

Harrison, R. M. and Yin, J.: Sources and processes affecting car-bonaceous aerosol in central England, Atmos. Environ., 42,1413–1423, doi:10.1016/j.atmosenv.2007.11.004, 2008.

Heal, M. R., Kumar, P., and Harrison, R. M.: Particles, airquality, policy and health, Chem. Soc. Rev., 41, 6606–6630,doi:10.1039/c2cs35076a, 2012.

Hodzic, A., Jimenez, J. L., Madronich, S., Canagaratna, M. R., De-Carlo, P. F., Kleinman, L., and Fast, J.: Modeling organic aerosolsin a megacity: potential contribution of semi-volatile and inter-mediate volatility primary organic compounds to secondary or-ganic aerosol formation, Atmos. Chem. Phys., 10, 5491–5514,doi:10.5194/acp-10-5491-2010, 2010.

Jathar, S. H., Farina, S. C., Robinson, A. L., and Adams, P. J.:The influence of semi-volatile and reactive primary emissionson the abundance and properties of global organic aerosol, At-mos. Chem. Phys., 11, 7727–7746, doi:10.5194/acp-11-7727-2011, 2011.

Jathar, S. H., Gordon, T. D., Hennigan, C. J., Pye, H. O. T.,Pouliot, G., Adams, P. J., Donahue, N. M., and Robinson, A. L.:Unspeciated organic emissions from combustion sources andtheir influence on the secondary organic aerosol budget in theUnited States, P. Natl. Acad. Sci. USA, 111, 10473–10478,doi:10.1073/pnas.1323740111, 2014.

Jimenez, J. L., Canagaratna, M. R., Donahue, N. M., Prevot, A. S.,Zhang, Q., Kroll, J. H., DeCarlo, P. F., Allan, J. D., Coe, H., Ng,N. L., Aiken, A. C., Docherty, K. S., Ulbrich, I. M., Grieshop,A. P., Robinson, A. L., Duplissy, J., Smith, J. D., Wilson, K. R.,Lanz, A. V., Hueglin, C., Sun, Y. L., Tian, J., Laaksonen, A.,Raatikainen, T., Rautiainen, J., Vaattovaara, P., Ehn, M., Kul-mala, M., Tomlinson, J. M., Collins, D. R., Cubison, M. J., Dun-lea, E. J., Huffman, J. A., Onasch, T. B., Alfarra, M. R., Williams,P. I., Bower, K., Kondo, Y., Schneider, J., Drewnick, F., Bor-rmann, S., Weimer, S., Demerjian, K., Salcedo, D., Cottrell, L.,Griffin, R., Takami, A., Miyoshi, T., Hatakeyama, S., Shimono,A., Sun, J. Y., Zhang, Y. M., Dzepina, K., Kimmel, J. R., Sueper,D., Jayne, J. T., Herndon, S. C., Trimborn, A. M., Williams, L. R.,Wood, E. C., Middlebrook, A. M., Kolb, C. E., Baltensperger, U.,and Worsnop, D. R.: Evolution of Organic Aerosols in the Atmo-sphere, Science, 326, 1525–1529, doi:10.1126/science.1180353,2009.

Koo, B., Knipping, E., and Yarwood, G.: 1.5-Dimensionalvolatility basis set approach for modeling organic aerosolin CAMx and CMAQ, Atmos. Environ., 95, 158–164,doi:10.1016/j.atmosenv.2014.06.031, 2014.

Kuenen, J. J. P., Visschedijk, A. J. H., Jozwicka, M., and Denier vander Gon, H. A. C.: TNO-MACC_II emission inventory; a multi-year (2003–2009) consistent high-resolution European emissioninventory for air quality modelling, Atmos. Chem. Phys., 14,10963–10976, doi:10.5194/acp-14-10963-2014, 2014.

Lane, T. E., Donahue, N. M., and Pandis, S. N.: Simulating sec-ondary organic aerosol formation using the volatility basis-setapproach in a chemical transport model, Atmos. Environ., 42,7439–7451, doi:10.1016/j.atmosenv.2008.06.026, 2008.


http://uk-air.defra.gov.uk/assets/documents/reports/cat15/1012131459_21897_Final_Report_291110.pdf

http://uk-air.defra.gov.uk/assets/documents/reports/cat15/1012131459_21897_Final_Report_291110.pdf

http://cdn.exxonmobil.com/~/media/global/files/outlook-for-energy/2015-outlook-for-energy_print-resolution.pdf











http://dx.doi.org/10.1073/pnas.1212272109

http://dx.doi.org/10.1021/es072476p


http://dx.doi.org/10.1016/j.atmosenv.2007.11.004

http://dx.doi.org/10.1039/c2cs35076a




http://dx.doi.org/10.1073/pnas.1323740111

http://dx.doi.org/10.1126/science.1180353





Lanz, V. A., Prévôt, A. S. H., Alfarra, M. R., Weimer, S., Mohr,C., DeCarlo, P. F., Gianini, M. F. D., Hueglin, C., Schneider, J.,Favez, O., D’Anna, B., George, C., and Baltensperger, U.: Char-acterization of aerosol chemical composition with aerosol massspectrometry in Central Europe: an overview, Atmos. Chem.Phys., 10, 10453–10471, doi:10.5194/acp-10-10453-2010, 2010.

Legates, D. R. and McCabe, G. J.: A refined index of modelperformance: a rejoinder, Int. J. Climatol., 33, 1053–1056,doi:10.1002/joc.3487, 2013.

Murphy, B. N. and Pandis, S. N.: Simulating the Formation ofSemivolatile Primary and Secondary Organic Aerosol in a Re-gional Chemical Transport Model, Environ. Sci. Technol., 43,4722–4728, doi:10.1021/Es803168a, 2009.

NAEI: UK Emission Mapping Methodology 2009, http://uk-air.defra.gov.uk/assets/documents/reports/cat07/1403100909_UK_Emission_Mapping_Methodology_2011-Issue_1.pdf (lastaccess: 24 May 2016), 2013.

NCAR: A Description of the Advanced Research WRF Version 3,Tech. rep., http://www2.mmm.ucar.edu/wrf/users/docs/arw_v3.pdf (last access: 24 May 2016), 2008.

NCAR: The NCAR Command Language (Version 6.3.0)[Software], UCAR/NCAR/CISL/TDD, Boulder, Colorado,doi:10.5065/D6WD3XH5, 2015.

NCEP: NCEP FNL Operational Model Global Tropospheric Anal-yses, continuing from July 1999, Research Data Archiveat the National Center for Atmospheric Research, Compu-tational and Information Systems Laboratory, Boulder, CO,doi:10.5065/D6M043C6, 2000.

Ng, N. L., Canagaratna, M. R., Zhang, Q., Jimenez, J. L., Tian,J., Ulbrich, I. M., Kroll, J. H., Docherty, K. S., Chhabra, P.S., Bahreini, R., Murphhy, S. M., Seinfeld, J. H., Hildebrandt,L., Donahue, N. M., DeCarlo, P. F., Lanz, V. A., Prévôt, A. S.H., Dinar, E., Rudich, Y., and Worsnop, D. R.: Organic aerosolcomponents observed in Northern Hemispheric datasets fromAerosol Mass Spectrometry, Atmos. Chem. Phys., 10, 4625–4641, doi:10.5194/acp-10-4625-2010, 2010.

Ng, N. L., Canagaratna, M. R., Jimenez, J. L., Zhang, Q., Ul-brich, I. M., and Worsnop, D. R.: Real-Time Methods for Esti-mating Organic Component Mass Concentrations from AerosolMass Spectrometer Data, Environ. Sci. Technol., 45, 910–916,doi:10.1021/es102951k, 2011.

Odum, J. R., Hoffmann, T., Bowman, F., Collins, D., Flagan, R. C.,and Seinfeld, J. H.: Gas/Particle Partitioning and Secondary Or-ganic Aerosol Yields, Environ. Sci. Technol., 30, 2580–2585,doi:10.1021/es950943+, 1996.

Paatero, P.: Least squares formulation of robust non-negativefactor analysis, Chemometr. Intel. Labor. Syst., 37, 23–35,doi:10.1016/S0169-7439(96)00044-5, 1997.

Paatero, P. and Tapper, U.: Positive matrix factorization: Anon-negative factor model with optimal utilization of er-ror estimates of data values, Environmetrics, 5, 111–126,doi:10.1002/env.3170050203, 1994.

Pan, X., Chin, M., Gautam, R., Bian, H., Kim, D., Colarco, P. R.,Diehl, T. L., Takemura, T., Pozzoli, L., Tsigaridis, K., Bauer,S., and Bellouin, N.: A multi-model evaluation of aerosols overSouth Asia: common problems and possible causes, Atmos.Chem. Phys., 15, 5903–5928, doi:10.5194/acp-15-5903-2015,2015.

Passant, N. R.: Speciation of UK emissions of non-methanevolatile organic compounds, Tech. rep., AEA TechnologyReport ENV-05452002, Culham, Abington, UK, availableat: https://uk-air.defra.gov.uk/assets/documents/reports/empire/AEAT_ENV_0545_final_v2.pdf (last access: 24 May 2016),2002.

Pöschl, U.: Atmospheric aerosols: Composition, transformation,climate and health effects, Angew. Chemie – Int. Edn., 44, 7520–7540, doi:10.1002/anie.200501122, 2005.

Presto, A. A., Miracolo, M. A., Donahue, N. M., and Robinson,A. L.: Secondary Organic Aerosol Formation from High-NOx

Photo-Oxidation of Low Volatility Precursors: n-Alkanes, Envi-ron. Sci. Technol., 44, 2029–2034, doi:10.1021/es903712r, 2010.

Putaud, J. P., Van Dingenen, R., Alastuey, A., Bauer, H., Birmili, W.,Cyrys, J., Flentje, H., Fuzzi, S., Gehrig, R., Hansson, H. C., Har-rison, R. M., Herrmann, H., Hitzenberger, R., Hüglin, C., Jones,A. M., Kasper-Giebl, A., Kiss, G., Kousa, A., Kuhlbusch, T. A. J.,Löschau, G., Maenhaut, W., Molnar, A., Moreno, T., Pekkanen,J., Perrino, C., Pitz, M., Puxbaum, H., Querol, X., Rodriguez, S.,Salma, I., Schwarz, J., Smolik, J., Schneider, J., Spindler, G., tenBrink, H., Tursic, J., Viana, M., Wiedensohler, A., and Raes, F.:A European aerosol phenomenology – 3: Physical and chemi-cal characteristics of particulate matter from 60 rural, urban, andkerbside sites across Europe, Atmos. Environ., 44, 1308–1320,doi:10.1016/j.atmosenv.2009.12.011, 2010.

Pye, H. O. T. and Seinfeld, J. H.: A global perspective on aerosolfrom low-volatility organic compounds, Atmos. Chem. Phys., 10,4377–4401, doi:10.5194/acp-10-4377-2010, 2010.

R Core Team: R: A Language and Environment for Statistical Com-puting, R Foundation for Statistical Computing, Vienna, Austria,http://www.R-project.org/ (last access: 24 May 2016), 2014.

Robinson, A. L., Donahue, N. M., Shrivastava, M. K., Weitkamp,E. A., Sage, A. M., Grieshop, A. P., Lane, T. E., Pierce, J. R.,and Pandis, S. N.: Rethinking organic aerosols: Semivolatileemissions and photochemical aging, Science, 315, 1259–1262,doi:10.1126/science.1133061, 2007.

Schauer, J. J., Kleeman, M. J., Cass, G. R., and Simoneit,B. R. T.: Measurement of Emissions from Air PollutionSources. 1. C1 through C29 Organic Compounds fromMeat Charbroiling, Environ. Sci. Technology, 33, 1566–1577,doi:10.1021/es980076j, 1999.

Schulz, M., Gauss, M., Benedictow, A., Jonson, J. E., Tsyro, S.,Nyiri, A., Simpson, D., Steensen, B. M., Klein, H., Valdebenito,A., Wind, P., Kirkevåg, A., Griesfeller, J., Bartnicki, J., Olivie,D., Grini, A., Iversen, T., Seland, ., Semeena, S. V., Fagerli,H., Aas, W., Hjellbrekke, A., Mareckova, K., Wankmuller, R.,Schneider, P., Solberg, S., Svendby, T., Liu, L., Posch, M., Vieno,M., Reis, S., Kryza, M., Werner, M., and Walaszek, K.: Trans-boundary Acidification, Eutrophication and Ground Level Ozonein Europe in 2011, Tech. rep., Norwegian Meteorological Insti-tute, Oslo, Norway, 1–205, 2013.

Shrivastava, M., Fast, J., Easter, R., Gustafson Jr., W. I., Zaveri, R.A., Jimenez, J. L., Saide, P., and Hodzic, A.: Modeling organicaerosols in a megacity: comparison of simple and complex rep-resentations of the volatility basis set approach, Atmos. Chem.Phys., 11, 6639–6662, doi:10.5194/acp-11-6639-2011, 2011.

Shrivastava, M., Easter, R. C., Liu, X., Zelenyuk, A., Singh, B.,Zhang, K., Ma, P., Chand, D., Ghan, S., Jimenez, J. L., Zhang, Q.,Fast, J., Rasch, P. J., and Tiitta, P.: Global transformation and fate



http://dx.doi.org/10.1002/joc.3487

http://dx.doi.org/10.1021/Es803168a

http://uk-air.defra.gov.uk/assets/documents/reports/cat07/1403100909_UK_Emission_Mapping_Methodology_2011-Issue_1.pdf



http://www2.mmm.ucar.edu/wrf/users/docs/arw_v3.pdf

http://www2.mmm.ucar.edu/wrf/users/docs/arw_v3.pdf

http://dx.doi.org/10.5065/D6WD3XH5

http://dx.doi.org/10.5065/D6M043C6


http://dx.doi.org/10.1021/es102951k

http://dx.doi.org/10.1021/es950943+

http://dx.doi.org/10.1016/S0169-7439(96)00044-5

http://dx.doi.org/10.1002/env.3170050203


https://uk-air.defra.gov.uk/assets/documents/reports/empire/AEAT_ENV_0545_final_v2.pdf

https://uk-air.defra.gov.uk/assets/documents/reports/empire/AEAT_ENV_0545_final_v2.pdf

http://dx.doi.org/10.1002/anie.200501122

http://dx.doi.org/10.1021/es903712r



http://www.R-project.org/

http://dx.doi.org/10.1126/science.1133061

http://dx.doi.org/10.1021/es980076j



of SOA: Implications of low-volatility SOA and gas-phase frag-mentation reactions, J. Geophys. Res.-Atmos., 120, 4169–4195,doi:10.1002/2014JD022563, 2015.

Shrivastava, M. K., Lane, T. E., Donahue, N. M., Pandis, S. N.,and Robinson, A. L.: Effects of gas particle partitioning andaging of primary emissions on urban and regional organicaerosol concentrations, J. Geophys. Res.-Atmos., 113, 1–16,doi:10.1029/2007jd009735, 2008.

Simpson, D., Yttri, K. E., Klimont, Z., Kupiainen, K., Caseiro, A.,Gelencser, A., Pio, C., Puxbaum, H., and Legrand, M.: Modelingcarbonaceous aerosol over Europe: Analysis of the CARBOSOLand EMEP EC/OC campaigns, J. Geophys. Res.-Atmos., 112, 1–26, doi:10.1029/2006jd008158, 2007.

Simpson, D., Benedictow, A., Berge, H., Bergström, R., Emberson,L. D., Fagerli, H., Flechard, C. R., Hayman, G. D., Gauss, M.,Jonson, J. E., Jenkin, M. E., Nyíri, A., Richter, C., Semeena, S.V., Tsyro, S., Tuovinen, J. P., Valdebenito, A., and Wind, P.: TheEMEP MSC-W chemical transport model – technical descrip-tion, Atmos. Chem. Phys., 12, 7825–7865, doi:10.5194/acp-12-7825-2012, 2012.

Spracklen, D. V., Jimenez, J. L., Carslaw, K. S., Worsnop, D. R.,Evans, M. J., Mann, G. W., Zhang, Q., Canagaratna, M. R.,Allan, J., Coe, H., McFiggans, G., Rap, A., and Forster, P.:Aerosol mass spectrometer constraint on the global secondaryorganic aerosol budget, Atmos. Chem. Phys., 11, 12109–12136,doi:10.5194/acp-11-12109-2011, 2011.

Tsigaridis, K., Daskalakis, N., Kanakidou, M., Adams, P. J., Ar-taxo, P., Bahadur, R., Balkanski, Y., Bauer, S. E., Bellouin, N.,Benedetti, A., Bergman, T., Berntsen, T. K., Beukes, J. P., Bian,H., Carslaw, K. S., Chin, M., Curci, G., Diehl, T., Easter, R.C., Ghan, S. J., Gong, S. L., Hodzic, A., Hoyle, C. R., Iversen,T., Jathar, S., Jimenez, J. L., Kaiser, J. W., Kirkevåg, A., Koch,D., Kokkola, H., Lee, Y. H., Lin, G., Liu, X., Luo, G., Ma, X.,Mann, G. W., Mihalopoulos, N., Morcrette, J.-J., Müller, J.-F.,Myhre, G., Myriokefalitakis, S., Ng, N. L., O’Donnell, D., Pen-ner, J. E., Pozzoli, L., Pringle, K. J., Russell, L. M., Schulz, M.,Sciare, J., Seland, Ø., Shindell, D. T., Sillman, S., Skeie, R. B.,Spracklen, D., Stavrakou, T., Steenrod, S. D., Takemura, T., Ti-itta, P., Tilmes, S., Tost, H., van Noije, T., van Zyl, P. G., vonSalzen, K., Yu, F., Wang, Z., Wang, Z., Zaveri, R. A., Zhang,H., Zhang, K., Zhang, Q., and Zhang, X.: The AeroCom eval-uation and intercomparison of organic aerosol in global mod-els, Atmos. Chem. Phys., 14, 10845–10895, doi:10.5194/acp-14-10845-2014, 2014.

Tsimpidi, A. P., Karydis, V. A., Zavala, M., Lei, W., Molina, L.,Ulbrich, I. M., Jimenez, J. L., and Pandis, S. N.: Evaluation of thevolatility basis-set approach for the simulation of organic aerosolformation in the Mexico City metropolitan area, Atmos. Chem.Phys., 10, 525–546, doi:10.5194/acp-10-525-2010, 2010.

Tsimpidi, A. P., Karydis, V. A., Pandis, S. N., and Lelieveld, J.:Global combustion sources of organic aerosols: Model compari-son with 84 AMS factor analysis data sets, Atmos. Chem. Phys.Discuss., doi:10.5194/acp-2015-989, in review, 2016.

Ulbrich, I. M., Canagaratna, M. R., Zhang, Q., Worsnop, D. R., andJimenez, J. L.: Interpretation of organic components from Posi-tive Matrix Factorization of aerosol mass spectrometric data, At-mos. Chem. Phys., 9, 2891–2918, doi:10.5194/acp-9-2891-2009,2009.

USEPA: Integrated Science Assessment for Particulate Matter(Final Report, http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=216546 (last access: 24 May 2016), 2009.

Vieno, M., Dore, A. J., Stevenson, D. S., Doherty, R., Heal, M. R.,Reis, S., Hallsworth, S., Tarrason, L., Wind, P., Fowler, D., Simp-son, D., and Sutton, M. A.: Modelling surface ozone during the2003 heat-wave in the UK, Atmos. Chem. and Phys., 10, 7963–7978, doi:10.5194/acp-10-7963-2010, 2010.

Vieno, M., Heal, M. R., Hallsworth, S., Famulari, D., Doherty, R.M., Dore, A. J., Tang, Y. S., Braban, C. F., Leaver, D., Sutton, M.A., and Reis, S.: The role of long-range transport and domes-tic emissions in determining atmospheric secondary inorganicparticle concentrations across the UK, Atmos. Chem. Phys., 14,8435–8447, doi:10.5194/acp-14-8435-2014, 2014.

Wickham, H.: ggplot2: Elegant Graphics for Data Analysis, ISBN978-0-387-98140-6, Springer-Verlag, New York, USA, availableat: http://ggplot2.org (last access: 24 May 2016), 2009.

Wiedinmyer, C., Akagi, S. K., Yokelson, R. J., Emmons, L. K., Al-Saadi, J. A., Orlando, J. J., and Soja, A. J.: The Fire INventoryfrom NCAR (FINN): a high resolution global model to estimatethe emissions from open burning, Geosci. Model Dev., 4, 625–641, doi:10.5194/gmd-4-625-2011, 2011.

Xu, L., Suresh, S., Guo, H., Weber, R. J., and Ng, N. L.: Aerosolcharacterization over the southeastern United States using high-resolution aerosol mass spectrometry: spatial and seasonal varia-tion of aerosol composition and sources with a focus on organicnitrates, Atmos. Chem. Phys., 15, 7307–7336, doi:10.5194/acp-15-7307-2015, 2015.

Xu, L., Williams, L. R., Young, D. E., Allan, J. D., Coe, H., Massoli,P., Fortner, E., Chhabra, P., Herndon, S., Brooks, W. A., Jayne, J.T., Worsnop, D. R., Aiken, A. C., Liu, S., Gorkowski, K., Dubey,M. K., Fleming, Z. L., Visser, S., Prévôt, A. S. H., and Ng, N. L.:Wintertime aerosol chemical composition, volatility, and spatialvariability in the greater London area, Atmos. Chem. Phys., 16,1139–1160, doi:10.5194/acp-16-1139-2016, 2016.

Young, D. E., Allan, J. D., Williams, P. I., Green, D. C., Flynn, M.J., Harrison, R. M., Yin, J., Gallagher, M. W., and Coe, H.: Inves-tigating the annual behaviour of submicron secondary inorganicand organic aerosols in London, Atmos. Chem. Phys., 15, 6351–6366, doi:10.5194/acp-15-6351-2015, 2015a.

Young, D. E., Allan, J. D., Williams, P. I., Green, D. C., Harri-son, R. M., Yin, J., Flynn, M. J., Gallagher, M. W., and Coe,H.: Investigating a two-component model of solid fuel organicaerosol in London: processes, PM1 contributions, and season-ality, Atmos. Chem. Phys., 15, 2429–2443, doi:10.5194/acp-15-2429-2015, 2015b.

Zhang, Q. J., Beekmann, M., Drewnick, F., Freutel, F., Schneider,J., Crippa, M., Prevot, A. S. H., Baltensperger, U., Poulain, L.,Wiedensohler, A., Sciare, J., Gros, V., Borbon, A., Colomb, A.,Michoud, V., Doussin, J., Denier van der Gon, H. A. C., Haeffe-lin, M., Dupont, J., Siour, G., Petetin, H., Bessagnet, B., Pan-dis, S. N., Hodzic, A., Sanchez, O., Honoré, C., and Perrus-sel, O.: Formation of organic aerosol in the Paris region duringthe MEGAPOLI summer campaign: evaluation of the volatility-basis-set approach within the CHIMERE model, Atmos. Chem.Phys., 13, 5767–5790, doi:10.5194/acp-13-5767-2013, 2013.


http://dx.doi.org/10.1002/2014JD022563

http://dx.doi.org/10.1029/2007jd009735

http://dx.doi.org/10.1029/2006jd008158







http://dx.doi.org/10.5194/acp-2015-989


http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=216546

http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=216546



http://ggplot2.org

http://dx.doi.org/10.5194/gmd-4-625-2011








simulating secondary organic aerosol from missing diesel ... · concluded that the contribution of...

Documents